{"meta": {"count": 457, "db_response_time_ms": 47, "page": 1, "per_page": 100, "groups_count": null, "x_query": {"oql": "get works where retracted is (false)\n  and year is (2025)\n  and title-abstract has ((stemmed \"artificial intelligence\" OR stemmed \"deep learning\" OR stemmed \"language model\" OR stemmed \"large language model\" OR stemmed \"machine learning\" OR stemmed \"neural network\" OR stemmed \"vision language model\") AND (stemmed \"data parallelism\" OR stemmed \"distributed training\" OR stemmed \"gradient communication\" OR stemmed \"model parallelism\" OR stemmed \"pipeline parallelism\" OR stemmed \"training checkpoint\"))\n  and type is (article or preprint or review);\nthen sample (457) of those works with seed (42)", "oqo": {"get_rows": "works", "filter_rows": [{"column_id": "is_retracted", "value": false}, {"column_id": "publication_year", "value": 2025}, {"join": "or", "filters": [{"column_id": "title_and_abstract.search", "value": "\"artificial intelligence\"", "operator": "has"}, {"column_id": "title_and_abstract.search", "value": "\"deep learning\"", "operator": "has"}, {"column_id": "title_and_abstract.search", "value": "\"language model\"", "operator": "has"}, {"column_id": "title_and_abstract.search", "value": "\"large language model\"", "operator": "has"}, {"column_id": "title_and_abstract.search", "value": "\"machine learning\"", "operator": "has"}, {"column_id": "title_and_abstract.search", "value": "\"neural network\"", "operator": "has"}, {"column_id": "title_and_abstract.search", "value": "\"vision language model\"", "operator": "has"}]}, {"join": "or", "filters": [{"column_id": "title_and_abstract.search", "value": "\"data parallelism\"", "operator": "has"}, {"column_id": "title_and_abstract.search", "value": "\"distributed training\"", "operator": "has"}, {"column_id": "title_and_abstract.search", "value": "\"gradient communication\"", "operator": "has"}, {"column_id": "title_and_abstract.search", "value": "\"model parallelism\"", "operator": "has"}, {"column_id": "title_and_abstract.search", "value": "\"pipeline parallelism\"", "operator": "has"}, {"column_id": "title_and_abstract.search", "value": "\"training checkpoint\"", "operator": "has"}]}, {"join": "or", "filters": [{"column_id": "type", "value": "article"}, {"column_id": "type", "value": "preprint"}, {"column_id": "type", "value": "review"}]}], "sample": 457, "seed": "42"}, "url": "/works?filter=is_retracted:false,publication_year:2025,title_and_abstract.search:%22artificial intelligence%22|%22deep learning%22|%22language model%22|%22large language model%22|%22machine learning%22|%22neural network%22|%22vision language model%22,title_and_abstract.search:%22data parallelism%22|%22distributed training%22|%22gradient communication%22|%22model parallelism%22|%22pipeline parallelism%22|%22training checkpoint%22,type:article|preprint|review&sample=457&seed=42&per_page=100&page=1"}, "cost_usd": 0.001}, "results": [{"id": "https://openalex.org/W4417351542", "doi": "https://doi.org/10.48550/arxiv.2506.10426", "title": "Towards Understanding Bugs in Distributed Training and Inference Frameworks for Large Language Models", "publication_year": 2025, "publication_date": "2025-06-12", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5100321702", "display_name": "Xiao Yu", "orcid": "https://orcid.org/0009-0001-0924-6432", "observed_orcids": ["https://orcid.org/0009-0001-0924-6432"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Yu, Xiao", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5103020390", "display_name": "Haoxuan Chen", "orcid": "https://orcid.org/0000-0002-8238-2764", "observed_orcids": ["https://orcid.org/0000-0002-8238-2764"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Chen, Haoxuan", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5065153631", "display_name": "Feifei Niu", "orcid": "https://orcid.org/0000-0002-4123-4554", "observed_orcids": ["https://orcid.org/0000-0002-4123-4554"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Niu, Feifei", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5047688641", "display_name": "Xing Hu", "orcid": "https://orcid.org/0000-0003-0093-3292", "observed_orcids": ["https://orcid.org/0000-0003-0093-3292"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Hu, Xing", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5051403641", "display_name": "Jacky Keung", "orcid": "https://orcid.org/0000-0002-3803-9600", "observed_orcids": ["https://orcid.org/0000-0002-3803-9600"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Keung, Jacky Wai", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5006669765", "display_name": "Xin Xia", "orcid": "https://orcid.org/0000-0002-6302-3256", "observed_orcids": ["https://orcid.org/0000-0002-6302-3256"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Xia, Xin", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"With": [0], "the": [1, 31, 65, 78, 114, 145, 205], "rapid": [2], "development": [3], "of": [4, 34, 67, 83, 117, 166, 207], "large": [5], "language": [6], "models": [7], "(LLMs),": [8], "distributed": [9, 90, 115, 131, 209], "training": [10, 22, 44, 210], "and": [11, 23, 49, 71, 95, 105, 108, 130, 135, 149, 156, 175, 211, 214, 230], "inference": [12, 24, 212], "frameworks": [13, 36, 119, 213], "like": [14, 144], "DeepSpeed": [15], "have": [16], "become": [17], "essential": [18], "for": [19, 61, 193, 203, 227], "scaling": [20], "model": [21], "across": [25, 87], "multiple": [26], "GPUs": [27], "or": [28, 158, 187], "nodes.": [29], "However,": [30], "increasing": [32], "complexity": [33], "these": [35, 118, 197], "brings": [37], "non-trivial": [38], "software": [39], "bugs,": [40], "which": [41], "may": [42], "degrade": [43], "performance,": [45], "cause": [46], "unexpected": [47], "failures,": [48], "result": [50], "in": [51], "significant": [52], "resource": [53], "waste.": [54], "Understanding": [55], "framework": [56], "bugs'": [57], "characteristics": [58], "is": [59], "fundamental": [60], "quality": [62], "assurance,": [63], "allowing": [64], "design": [66], "more": [68], "effective": [69], "debugging": [70, 229], "repair": [72], "methods.": [73], "Thus,": [74], "our": [75], "paper": [76], "conducts": [77], "first": [79], "large-scale": [80], "empirical": [81], "analysis": [82], "308": [84], "fixed": [85], "bugs": [86, 138], "three": [88], "popular": [89], "training/inference": [91], "frameworks:": [92], "DeepSpeed,": [93], "Megatron-LM,": [94], "Colossal-AI.": [96], "We": [97], "examine": [98], "bug": [99, 103, 122, 153, 167], "symptoms,": [100], "root": [101, 123, 150], "causes,": [102, 124, 151], "identification": [104], "fixing": [106, 111, 136], "efforts,": [107], "common": [109], "low-effort": [110], "strategies.": [112], "Additionally,": [113], "nature": [116], "introduces": [120], "unique": [121], "such": [125, 179], "as": [126, 180], "allocation": [127], "strategy": [128], "error": [129], "communication": [132], "error.": [133], "Diagnosing": [134], "complex": [137], "remains": [139], "challenging": [140], "due": [141], "to": [142, 223], "factors": [143], "disconnect": [146], "between": [147], "symptoms": [148], "high": [152], "reproduction": [154], "costs,": [155], "low-level": [157], "cross-component": [159], "interactions.": [160], "Interestingly,": [161], "we": [162, 199], "observe": [163], "that": [164], "48%": [165], "fixes": [168], "require": [169], "minimal": [170], "code": [171], "changes": [172], "(<=10": [173], "LOC)": [174], "follow": [176], "simple": [177], "strategies": [178], "conditional": [181], "logic": [182], "optimization,": [183], "parameter": [184], "handling": [185], "enhancement,": [186], "version": [188], "compatibility": [189], "handling,": [190], "indicating": [191], "potential": [192], "automation.": [194], "Based": [195], "on": [196], "insights,": [198], "offer": [200], "several": [201], "implications": [202], "improving": [204], "reliability": [206], "both": [208], "their": [215], "dependent": [216], "LLM": [217], "projects,": [218], "while": [219], "also": [220], "identifying": [221], "opportunities": [222], "leverage": [224], "LLM-based": [225], "tools": [226], "automated": [228], "repair.": [231]}, "primary_topic": {"id": "https://openalex.org/T10743", "display_name": "Software Testing and Debugging Techniques", "score": 0.9398999810218811, "subfield": {"id": "https://openalex.org/subfields/1712", "display_name": "Software"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10743", "display_name": "Software Testing and Debugging Techniques", "score": 0.9398999810218811, "subfield": {"id": "https://openalex.org/subfields/1712", "display_name": "Software"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10260", "display_name": "Software Engineering Research", "score": 0.05640000104904175, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T12127", "display_name": "Software System Performance and Reliability", "score": 0.0017999999690800905, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2506.10426", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2506.10426", "pdf_url": "https://arxiv.org/pdf/2506.10426", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4406863004", "doi": "https://doi.org/10.2139/ssrn.5041028", "title": "Distributed Learning Architectures", "publication_year": 2025, "publication_date": "2025-01-01", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5049302670", "display_name": "Elizabeth Henry", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": true, "raw_author_name": "Elizabeth Henry", "raw_affiliation_strings": ["Independent"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Independent", "institution_ids": []}]}, {"author_position": "last", "author": {"id": null, "display_name": "Edwin Frank", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Edwin Frank", "raw_affiliation_strings": ["Independent"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Independent", "institution_ids": []}]}], "abstract_inverted_index": null, "primary_topic": {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.6327999830245972, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.6327999830245972, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.23180000483989716, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10249", "display_name": "Distributed Control Multi-Agent Systems", "score": 0.04769999906420708, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.2139/ssrn.5041028", "is_oa": true, "landing_page_url": "https://doi.org/10.2139/ssrn.5041028", "pdf_url": null, "source": {"id": "https://openalex.org/S4210172589", "display_name": "SSRN Electronic Journal", "issn_l": "1556-5068", "issn": ["1556-5068"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I1318003438", "host_organization_name": "RELX Group (Netherlands)", "host_organization_lineage": ["https://openalex.org/I1318003438"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "acceptedVersion", "is_accepted": true, "is_published": false, "raw_source_name": null, "raw_type": "posted-content"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4411992388", "doi": "https://doi.org/10.1016/j.neucom.2025.130835", "title": "Towards bandwidth efficient edge\u2013cloud collaborative deep learning with Data Importance driven Compression", "publication_year": 2025, "publication_date": "2025-07-03", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Yalin Jiang", "orcid": "https://orcid.org/0000-0001-6078-6402", "observed_orcids": ["https://orcid.org/0000-0001-6078-6402"]}, "institutions": [{"id": "https://openalex.org/I87445476", "display_name": "Xi'an Jiaotong University", "ror": "https://ror.org/017zhmm22", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I87445476"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yalin Jiang", "raw_affiliation_strings": ["School of Computer Science and Technology, Xi\u2019an Jiaotong University, Xi\u2019an, 710049, China"], "raw_orcid": "https://orcid.org/0000-0001-6078-6402", "affiliations": [{"raw_affiliation_string": "School of Computer Science and Technology, Xi\u2019an Jiaotong University, Xi\u2019an, 710049, China", "institution_ids": ["https://openalex.org/I87445476"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5010106841", "display_name": "Peng Fei Zhao", "orcid": "https://orcid.org/0000-0001-7033-9315", "observed_orcids": ["https://orcid.org/0000-0001-7033-9315"]}, "institutions": [{"id": "https://openalex.org/I87445476", "display_name": "Xi'an Jiaotong University", "ror": "https://ror.org/017zhmm22", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I87445476"]}], "countries": ["CN"], "is_corresponding": true, "raw_author_name": "Peng Zhao", "raw_affiliation_strings": ["National Engineering Laboratory for Big Data Analytics, Xi\u2019an, 710049, China", "School of Mathematics and Statistics, Xi\u2019an Jiaotong University, Xi\u2019an, 710049, China"], "raw_orcid": "https://orcid.org/0000-0001-7033-9315", "affiliations": [{"raw_affiliation_string": "National Engineering Laboratory for Big Data Analytics, Xi\u2019an, 710049, China", "institution_ids": []}, {"raw_affiliation_string": "School of Mathematics and Statistics, Xi\u2019an Jiaotong University, Xi\u2019an, 710049, China", "institution_ids": ["https://openalex.org/I87445476"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5020695106", "display_name": "Cong Zhao", "orcid": "https://orcid.org/0009-0005-1356-0431", "observed_orcids": ["https://orcid.org/0009-0005-1356-0431", "https://orcid.org/0000-0002-9002-4350"]}, "institutions": [{"id": "https://openalex.org/I87445476", "display_name": "Xi'an Jiaotong University", "ror": "https://ror.org/017zhmm22", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I87445476"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Cong Zhao", "raw_affiliation_strings": ["National Engineering Laboratory for Big Data Analytics, Xi\u2019an, 710049, China", "School of Mathematics and Statistics, Xi\u2019an Jiaotong University, Xi\u2019an, 710049, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "National Engineering Laboratory for Big Data Analytics, Xi\u2019an, 710049, China", "institution_ids": []}, {"raw_affiliation_string": "School of Mathematics and Statistics, Xi\u2019an Jiaotong University, Xi\u2019an, 710049, China", "institution_ids": ["https://openalex.org/I87445476"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5100876534", "display_name": "Jie Lin", "orcid": "https://orcid.org/0000-0003-1785-8137", "observed_orcids": ["https://orcid.org/0000-0003-1785-8137"]}, "institutions": [{"id": "https://openalex.org/I87445476", "display_name": "Xi'an Jiaotong University", "ror": "https://ror.org/017zhmm22", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I87445476"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Jie Lin", "raw_affiliation_strings": ["School of Computer Science and Technology, Xi\u2019an Jiaotong University, Xi\u2019an, 710049, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Computer Science and Technology, Xi\u2019an Jiaotong University, Xi\u2019an, 710049, China", "institution_ids": ["https://openalex.org/I87445476"]}]}], "abstract_inverted_index": null, "primary_topic": {"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.9761000275611877, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.9761000275611877, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.012900000438094139, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.0017999999690800905, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1016/j.neucom.2025.130835", "is_oa": false, "landing_page_url": "https://doi.org/10.1016/j.neucom.2025.130835", "pdf_url": null, "source": {"id": "https://openalex.org/S45693802", "display_name": "Neurocomputing", "issn_l": "0925-2312", "issn": ["0925-2312", "1872-8286"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "doyens", "jufo-2", "ki-jl-2", "norway-2"], "host_organization": "https://openalex.org/P4310320990", "host_organization_name": "Elsevier BV", "host_organization_lineage": ["https://openalex.org/P4310320990"], "host_organization_lineage_names": ["Elsevier BV"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Neurocomputing", "raw_type": "journal-article"}, "cited_by_count": 4, "is_retracted": false}, {"id": "https://openalex.org/W7104290530", "doi": "https://doi.org/10.52783/jisem.v10i60s.13192", "title": "Privacy Preservation in Data-Hungry Deep Learning: A Comprehensive Review of Attacks and Techniques", "publication_year": 2025, "publication_date": "2025-09-30", "type": "article", "language": null, "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Omar Nassim Adel Benyamina", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": true, "raw_author_name": "Omar Nassim Adel Benyamina", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Deep": [0], "learning": [1, 29, 55, 81], "(DL)": [2], "has": [3, 31], "achieved": [4], "remarkable": [5], "success": [6], "across": [7], "various": [8], "domains,": [9], "including": [10], "healthcare,": [11], "finance,": [12], "and": [13, 45, 73, 99, 102, 113, 138, 153, 171, 189], "natural": [14], "language": [15], "processing;": [16], "however,": [17], "its": [18], "reliance": [19], "on": [20, 59], "sensitive": [21], "data": [22], "poses": [23], "significant": [24], "privacy": [25, 43, 71], "risks.": [26], "Privacy-preserving": [27], "deep": [28, 54], "(PPDL)": [30, 56], "therefore": [32], "emerged": [33], "as": [34, 83, 107], "a": [35], "critical": [36], "research": [37], "direction,": [38], "integrating": [39], "cryptographic": [40], "techniques,": [41], "statistical": [42], "mechanisms,": [44], "distributed": [46], "training": [47], "paradigms.": [48], "This": [49], "survey": [50], "reviews": [51], "state-of-the-art": [52], "privacy-preserving": [53], "techniques": [57, 90], "centered": [58], "homomorphic": [60], "encryption": [61], "(HE),": [62], "secure": [63, 74], "multi-party": [64], "computation": [65], "(SMPC)": [66], "(and": [67], "hybrid": [68], "protocols),": [69], "differential": [70, 136], "(DP),": [72], "enclaves": [75], "(SE/TEEs).": [76], "We": [77, 93, 180], "also": [78], "position": [79], "federated": [80], "(FL)": [82], "an": [84], "orchestration": [85], "paradigm": [86, 147], "that": [87, 144, 182], "composes": [88], "these": [89], "at": [91, 166], "scale.": [92], "systematically": [94], "analyze": [95], "their": [96], "efficacy,": [97], "privacy,": [98, 137], "efficiency": [100], "trade-offs,": [101], "map": [103], "common": [104], "attack": [105], "vectors\u2014such": [106], "reconstruction,": [108], "inversion,": [109], "membership": [110], "inference,": [111], "poisoning,": [112], "hardware-level": [114], "side": [115], "channels\u2014to": [116], "representative": [117], "defenses.": [118], "Bibliometric": [119], "analysis": [120], "using": [121], "VOSviewer": [122], "further": [123], "highlights": [124], "the": [125, 129, 167, 193], "thematic": [126], "structure": [127], "of": [128, 169], "field,": [130], "with": [131], "strong": [132, 156], "clusters": [133], "around": [134], "cryptography,": [135, 188], "system-level": [139], "optimization.": [140], "Our": [141], "findings": [142], "reveal": [143], "no": [145], "single": [146], "suffices": [148], "in": [149], "practice:": [150], "while": [151], "HE": [152], "SMPC": [154], "provide": [155], "confidentiality,": [157], "they": [158], "incur": [159], "high": [160], "costs;": [161], "DP": [162], "enables": [163], "formal": [164], "guarantees": [165], "expense": [168], "accuracy;": [170], "FL": [172], "reduces": [173], "raw-data": [174], "exposure": [175], "but": [176], "introduces": [177], "novel": [178], "vulnerabilities.": [179], "conclude": [181], "hybrid,": [183], "layered": [184], "strategies": [185], "combining": [186], "DP,": [187], "robust": [190], "aggregation": [191], "are": [192], "most": [194], "promising": [195], "path": [196], "toward": [197], "scalable,": [198], "trustworthy": [199], "PPDL": [200], "for": [201], "real-world": [202], "deployment.": [203]}, "primary_topic": {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.9984999895095825, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.9984999895095825, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10237", "display_name": "Cryptography and Data Security", "score": 0.0010000000474974513, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11614", "display_name": "Cloud Data Security Solutions", "score": 9.999999747378752e-05, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.52783/jisem.v10i60s.13192", "is_oa": true, "landing_page_url": "https://doi.org/10.52783/jisem.v10i60s.13192", "pdf_url": "https://jisem-journal.com/index.php/journal/article/download/13192/6175", "source": {"id": "https://openalex.org/S2737791640", "display_name": "Journal of Information Systems Engineering & Management", "issn_l": "2468-4376", "issn": ["2468-4376", "2468-2071"], "is_oa": true, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core"], "host_organization": "https://openalex.org/P4310320063", "host_organization_name": "Lectito Journals", "host_organization_lineage": ["https://openalex.org/P4310320063"], "host_organization_lineage_names": ["Lectito Journals"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Journal of Information Systems Engineering and Management", "raw_type": "journal-article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4414116822", "doi": "https://doi.org/10.1109/tpds.2025.3609152", "title": "XDGNN: Efficient Distributed GNN Training via Explanation-Guided Subgraph Expansion", "publication_year": 2025, "publication_date": "2025-09-11", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Jie Gao", "orcid": "https://orcid.org/0000-0002-3542-5389", "observed_orcids": ["https://orcid.org/0000-0002-3542-5389"]}, "institutions": [{"id": "https://openalex.org/I23923803", "display_name": "University of Exeter", "ror": "https://ror.org/03yghzc09", "country_code": "GB", "type": "education", "lineage": ["https://openalex.org/I23923803"]}], "countries": ["GB"], "is_corresponding": false, "raw_author_name": "Jie Gao", "raw_affiliation_strings": ["Department of Computer Science, University of Exeter, Exeter, U.K"], "raw_orcid": "https://orcid.org/0000-0002-3542-5389", "affiliations": [{"raw_affiliation_string": "Department of Computer Science, University of Exeter, Exeter, U.K", "institution_ids": ["https://openalex.org/I23923803"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5002896600", "display_name": "Jia Wen Hu", "orcid": "https://orcid.org/0000-0001-5406-8420", "observed_orcids": ["https://orcid.org/0000-0001-5406-8420"]}, "institutions": [{"id": "https://openalex.org/I23923803", "display_name": "University of Exeter", "ror": "https://ror.org/03yghzc09", "country_code": "GB", "type": "education", "lineage": ["https://openalex.org/I23923803"]}], "countries": ["GB"], "is_corresponding": false, "raw_author_name": "Jia Hu", "raw_affiliation_strings": ["Department of Computer Science, University of Exeter, Exeter, U.K"], "raw_orcid": "https://orcid.org/0000-0001-5406-8420", "affiliations": [{"raw_affiliation_string": "Department of Computer Science, University of Exeter, Exeter, U.K", "institution_ids": ["https://openalex.org/I23923803"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100770003", "display_name": "Geyong Min", "orcid": "https://orcid.org/0000-0003-1395-7314", "observed_orcids": ["https://orcid.org/0000-0003-1395-7314"]}, "institutions": [{"id": "https://openalex.org/I23923803", "display_name": "University of Exeter", "ror": "https://ror.org/03yghzc09", "country_code": "GB", "type": "education", "lineage": ["https://openalex.org/I23923803"]}], "countries": ["GB"], "is_corresponding": false, "raw_author_name": "Geyong Min", "raw_affiliation_strings": ["Department of Computer Science, University of Exeter, Exeter, U.K"], "raw_orcid": "https://orcid.org/0000-0003-1395-7314", "affiliations": [{"raw_affiliation_string": "Department of Computer Science, University of Exeter, Exeter, U.K", "institution_ids": ["https://openalex.org/I23923803"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5002241280", "display_name": "Fei Hao", "orcid": "https://orcid.org/0000-0001-5288-5523", "observed_orcids": ["https://orcid.org/0000-0001-5288-5523"]}, "institutions": [{"id": "https://openalex.org/I88830068", "display_name": "Shaanxi Normal University", "ror": "https://ror.org/0170z8493", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I88830068"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Fei Hao", "raw_affiliation_strings": ["School of Artificial Intelligence and Computer Science, Shaanxi Normal University, Xi&#x2019;an, China", "School of Artificial Intelligence and Computer Science, Shaanxi Normal University, Xi&#x0027;an, China"], "raw_orcid": "https://orcid.org/0000-0001-5288-5523", "affiliations": [{"raw_affiliation_string": "School of Artificial Intelligence and Computer Science, Shaanxi Normal University, Xi&#x2019;an, China", "institution_ids": ["https://openalex.org/I88830068"]}, {"raw_affiliation_string": "School of Artificial Intelligence and Computer Science, Shaanxi Normal University, Xi&#x0027;an, China", "institution_ids": ["https://openalex.org/I88830068"]}]}], "abstract_inverted_index": {"Graph": [0], "neural": [1], "network": [2], "(GNN)": [3], "is": [4, 21, 43], "a": [5, 53, 96], "state-of-the-art": [6], "technique": [7, 119], "for": [8, 40], "learning": [9], "structural": [10], "information": [11, 134], "from": [12], "graph": [13, 138], "data.": [14], "However,": [15], "training": [16, 45, 69, 87, 100, 145, 151, 171, 183], "GNNs": [17, 42], "on": [18, 146], "large-scale": [19], "graphs": [20, 30], "very": [22], "challenging": [23], "due": [24], "to": [25, 60, 80], "the": [26, 32, 62, 86, 104, 152, 161, 175], "size": [27], "of": [28, 35], "real-world": [29], "and": [31, 73, 84, 108], "message-passing": [33], "architecture": [34], "GNNs.": [36], "One": [37], "promising": [38], "approach": [39], "scaling": [41], "distributed": [44, 67, 98, 144, 181], "across": [46], "multiple": [47], "accelerators,": [48], "where": [49], "each": [50], "accelerator": [51], "holds": [52], "partitioned": [54], "subgraph": [55, 117], "that": [56, 102, 120, 167], "fits": [57], "in": [58, 64, 154, 160], "memory": [59], "train": [61], "model": [63, 153, 176], "parallel.": [65], "Existing": [66], "GNN": [68, 99, 182], "methods": [70, 129], "require": [71], "frequent": [72], "prohibitive": [74], "embedding": [75], "exchanges": [76], "between": [77], "partitions,": [78, 132], "leading": [79], "substantial": [81], "communication": [82, 106], "overhead": [83], "limited": [85], "efficiency.": [88], "To": [89], "address": [90], "this": [91], "challenge,": [92], "we": [93, 113], "propose": [94], "XDGNN,": [95], "novel": [97], "method": [101], "eliminates": [103], "forward": [105, 162], "bottleneck": [107], "thus": [109], "accelerates": [110], "training.": [111], "Specifically,": [112], "design": [114], "an": [115], "explanation-guided": [116], "expansion": [118], "incorporates": [121], "important": [122], "structures": [123], "identified": [124], "by": [125, 137], "eXplanation": [126], "AI": [127], "(XAI)": [128], "into": [130], "local": [131], "mitigating": [133], "loss": [135], "caused": [136], "partitioning.": [139], "Then,": [140], "XDGNN": [141, 168], "conducts": [142], "communication-free": [143], "these": [147], "self-contained": [148], "partitions": [149], "through": [150], "parallel": [155], "without": [156], "communicating": [157], "node": [158], "embeddings": [159], "phase.": [163], "Extensive": [164], "experiments": [165], "demonstrate": [166], "significantly": [169], "improves": [170], "efficiency": [172], "while": [173], "maintaining": [174], "accuracy": [177], "compared": [178], "with": [179], "current": [180], "methods.": [184]}, "primary_topic": {"id": "https://openalex.org/T11273", "display_name": "Advanced Graph Neural Networks", "score": 0.9423999786376953, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T11273", "display_name": "Advanced Graph Neural Networks", "score": 0.9423999786376953, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T12292", "display_name": "Graph Theory and Algorithms", "score": 0.05299999937415123, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.0006000000284984708, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/tpds.2025.3609152", "is_oa": false, "landing_page_url": "https://doi.org/10.1109/tpds.2025.3609152", "pdf_url": null, "source": {"id": "https://openalex.org/S97130795", "display_name": "IEEE Transactions on Parallel and Distributed Systems", "issn_l": "1045-9219", "issn": ["1045-9219", "1558-2183", "2161-9883"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-2", "norway-2"], "host_organization": "https://openalex.org/P4310319808", "host_organization_name": "Institute of Electrical and Electronics Engineers", "host_organization_lineage": ["https://openalex.org/P4310319808"], "host_organization_lineage_names": ["Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Transactions on Parallel and Distributed Systems", "raw_type": "journal-article"}, "cited_by_count": 1, "is_retracted": false}, {"id": "https://openalex.org/W4416047179", "doi": "https://doi.org/10.48047/jocaaa.2025.34.7.23", "title": "Evaluating Infrastructure Bottlenecks in Distributed AI Training: A Performance Benchmark Study", "publication_year": 2025, "publication_date": "2025-10-25", "type": "article", "language": null, "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Sayantan Ghosh", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": true, "raw_author_name": "Sayantan Ghosh", "raw_affiliation_strings": ["Infra Engineering Leader"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Infra Engineering Leader", "institution_ids": []}]}], "abstract_inverted_index": {"The": [0], "rapid": [1], "expansion": [2], "of": [3, 11], "artificial": [4], "intelligence": [5], "(AI)": [6], "and": [7], "the": [8, 16], "increasing": [9], "complexity": [10], "deeplearning": [12], "models": [13], "have": [14], "intensified": [15], "need": [17], "for": [18], "efficient": [19], "distributed": [20], "training": [21], "systems.": [22], "This": [23], "study,": [24], "titled": [25], "\u201cEvaluating": [26], "Infrastructure": [27]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.5717999935150146, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.5717999935150146, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.2062000036239624, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10715", "display_name": "Distributed and Parallel Computing Systems", "score": 0.06270000338554382, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.48047/jocaaa.2025.34.7.23", "is_oa": true, "landing_page_url": "https://doi.org/10.48047/jocaaa.2025.34.7.23", "pdf_url": "https://www.eudoxuspress.com/index.php/pub/article/download/4054/2944", "source": {"id": "https://openalex.org/S23262349", "display_name": "Journal of Computational Analysis and Applications", "issn_l": "1521-1398", "issn": ["1521-1398", "1572-9206"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": ["jufo-1", "norway-1"], "host_organization": "https://openalex.org/P4310319900", "host_organization_name": "Springer Science+Business Media", "host_organization_lineage": ["https://openalex.org/P4310319900", "https://openalex.org/P4310319965"], "host_organization_lineage_names": ["Springer Science+Business Media", "Springer Nature"], "type": "journal"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Journal of Computational Analysis and Applications", "raw_type": "journal-article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4407958854", "doi": "https://doi.org/10.1177/10943420251317994", "title": "Performance of linear solvers in tensor-train format on current multicore architectures", "publication_year": 2025, "publication_date": "2025-02-26", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5048878210", "display_name": "Melven R\u00f6hrig-Z\u00f6llner", "orcid": "https://orcid.org/0000-0001-9851-5886", "observed_orcids": ["https://orcid.org/0000-0001-9851-5886"]}, "institutions": [{"id": "https://openalex.org/I2898391981", "display_name": "Deutsches Zentrum f\u00fcr Luft- und Raumfahrt e. V. (DLR)", "ror": "https://ror.org/04bwf3e34", "country_code": "DE", "type": "facility", "lineage": ["https://openalex.org/I1305996414", "https://openalex.org/I2898391981"]}], "countries": ["DE"], "is_corresponding": false, "raw_author_name": "Melven R\u00f6hrig-Z\u00f6llner", "raw_affiliation_strings": ["German Aerospace Center (DLR)", "German Aerospace Center (DLR), Institute of Software Technology, Cologne, Germany"], "raw_orcid": "https://orcid.org/0000-0001-9851-5886", "affiliations": [{"raw_affiliation_string": "German Aerospace Center (DLR)", "institution_ids": ["https://openalex.org/I2898391981"]}, {"raw_affiliation_string": "German Aerospace Center (DLR), Institute of Software Technology, Cologne, Germany", "institution_ids": ["https://openalex.org/I2898391981"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5093497301", "display_name": "Manuel Joey Becklas", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I2898391981", "display_name": "Deutsches Zentrum f\u00fcr Luft- und Raumfahrt e. V. (DLR)", "ror": "https://ror.org/04bwf3e34", "country_code": "DE", "type": "facility", "lineage": ["https://openalex.org/I1305996414", "https://openalex.org/I2898391981"]}], "countries": ["DE"], "is_corresponding": false, "raw_author_name": "Manuel Becklas", "raw_affiliation_strings": ["German Aerospace Center (DLR)", "German Aerospace Center (DLR), Institute of Software Technology, Cologne, Germany"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "German Aerospace Center (DLR)", "institution_ids": ["https://openalex.org/I2898391981"]}, {"raw_affiliation_string": "German Aerospace Center (DLR), Institute of Software Technology, Cologne, Germany", "institution_ids": ["https://openalex.org/I2898391981"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5010377968", "display_name": "Jonas Thies", "orcid": "https://orcid.org/0000-0001-9231-9999", "observed_orcids": ["https://orcid.org/0000-0001-9231-9999"]}, "institutions": [{"id": "https://openalex.org/I98358874", "display_name": "Delft University of Technology", "ror": "https://ror.org/02e2c7k09", "country_code": "NL", "type": "education", "lineage": ["https://openalex.org/I98358874"]}], "countries": ["NL"], "is_corresponding": false, "raw_author_name": "Jonas Thies", "raw_affiliation_strings": ["Delft University of Technology", "Delft Institute of Applied Mathematics, Delft University of Technology, Delft, Netherlands"], "raw_orcid": "https://orcid.org/0000-0001-9231-9999", "affiliations": [{"raw_affiliation_string": "Delft University of Technology", "institution_ids": ["https://openalex.org/I98358874"]}, {"raw_affiliation_string": "Delft Institute of Applied Mathematics, Delft University of Technology, Delft, Netherlands", "institution_ids": ["https://openalex.org/I98358874"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5070137241", "display_name": "Achim Basermann", "orcid": "https://orcid.org/0000-0003-3637-3231", "observed_orcids": ["https://orcid.org/0000-0003-3637-3231"]}, "institutions": [{"id": "https://openalex.org/I2898391981", "display_name": "Deutsches Zentrum f\u00fcr Luft- und Raumfahrt e. V. (DLR)", "ror": "https://ror.org/04bwf3e34", "country_code": "DE", "type": "facility", "lineage": ["https://openalex.org/I1305996414", "https://openalex.org/I2898391981"]}], "countries": ["DE"], "is_corresponding": false, "raw_author_name": "Achim Basermann", "raw_affiliation_strings": ["German Aerospace Center (DLR)", "German Aerospace Center (DLR), Institute of Software Technology, Cologne, Germany"], "raw_orcid": "https://orcid.org/0000-0003-3637-3231", "affiliations": [{"raw_affiliation_string": "German Aerospace Center (DLR)", "institution_ids": ["https://openalex.org/I2898391981"]}, {"raw_affiliation_string": "German Aerospace Center (DLR), Institute of Software Technology, Cologne, Germany", "institution_ids": ["https://openalex.org/I2898391981"]}]}], "abstract_inverted_index": {"Tensor": [0], "networks": [1], "are": [2, 17, 45], "a": [3, 117, 166, 171, 182, 195], "class": [4], "of": [5, 13, 23, 82, 115, 143, 148, 174], "algorithms": [6, 36], "aimed": [7], "at": [8], "reducing": [9], "the": [10, 63, 80, 89, 113, 131, 141, 175, 186, 190], "computational": [11, 67, 109], "complexity": [12, 110], "high-dimensional": [14, 119], "problems.": [15], "They": [16], "used": [18], "in": [19, 34, 88], "an": [20], "increasing": [21], "number": [22, 142], "applications,": [24], "from": [25], "quantum": [26], "simulations": [27], "to": [28, 39, 48, 133], "machine": [29], "learning.": [30], "Exploiting": [31], "data": [32], "parallelism": [33], "these": [35], "is": [37], "key": [38], "using": [40], "modern": [41], "hardware.": [42], "However,": [43], "there": [44], "several": [46], "ways": [47], "map": [49], "required": [50], "tensor": [51, 160], "operations": [52, 145], "onto": [53], "linear": [54, 86, 176], "algebra": [55], "routines": [56], "(\u201cbuilding": [57], "blocks\u201d).": [58], "Optimizing": [59], "this": [60, 76], "mapping": [61], "impacts": [62], "numerical": [64, 69], "behavior,": [65], "so": [66], "and": [68, 104, 138, 159], "aspects": [70], "must": [71], "be": [72], "considered": [73], "hand-in-hand.": [74], "In": [75, 162], "paper": [77], "we": [78, 164, 179], "discuss": [79], "performance": [81], "solvers": [83], "for": [84, 122, 136, 189], "low-rank": [85], "systems": [87], "tensor-train": [90], "format": [91], "(also": [92], "known": [93], "as": [94], "matrix-product": [95], "states).": [96], "We": [97, 106, 150], "consider": [98], "three": [99], "popular": [100], "algorithms:": [101], "TT-GMRES,": [102], "MALS,": [103], "AMEn.": [105], "illustrate": [107], "their": [108], "based": [111, 169], "on": [112, 170, 194], "example": [114], "discretizing": [116], "simple": [118], "PDE": [120], "in,": [121], "example,": [123], "50": [124], "10": [125], "grid": [126], "points.": [127], "This": [128], "shows": [129], "that": [130], "projection": [132], "smaller": [134], "sub-problems": [135], "MALS": [137], "AMEn": [139], "reduces": [140], "floating-point": [144], "by": [146], "orders": [147], "magnitude.": [149], "suggest": [151], "optimizations": [152], "regarding": [153], "orthogonalization": [154], "steps,": [155], "singular": [156], "value": [157], "decompositions,": [158], "contractions.": [161], "addition,": [163], "propose": [165], "generic": [167], "preconditioner": [168], "TT-rank-1": [172], "approximation": [173], "operator.": [177], "Overall,": [178], "obtain": [180], "roughly": [181], "5\u00d7": [183], "speedup": [184], "over": [185], "reference": [187], "algorithm": [188], "fastest": [191], "method": [192], "(AMEn)": [193], "current": [196], "multicore": [197], "CPU.": [198]}, "primary_topic": {"id": "https://openalex.org/T12303", "display_name": "Tensor decomposition and applications", "score": 0.8208000063896179, "subfield": {"id": "https://openalex.org/subfields/2605", "display_name": "Computational Mathematics"}, "field": {"id": "https://openalex.org/fields/26", "display_name": "Mathematics"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T12303", "display_name": "Tensor decomposition and applications", "score": 0.8208000063896179, "subfield": {"id": "https://openalex.org/subfields/2605", "display_name": "Computational Mathematics"}, "field": {"id": "https://openalex.org/fields/26", "display_name": "Mathematics"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10792", "display_name": "Matrix Theory and Algorithms", "score": 0.13269999623298645, "subfield": {"id": "https://openalex.org/subfields/1703", "display_name": "Computational Theory and Mathematics"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.01549999974668026, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1177/10943420251317994", "is_oa": true, "landing_page_url": "https://doi.org/10.1177/10943420251317994", "pdf_url": null, "source": {"id": "https://openalex.org/S60606485", "display_name": "The International Journal of High Performance Computing Applications", "issn_l": "0890-2720", "issn": ["0890-2720", "1094-3420", "1741-2846"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-2", "norway-2"], "host_organization": "https://openalex.org/P4310320017", "host_organization_name": "SAGE Publishing", "host_organization_lineage": ["https://openalex.org/P4310320017"], "host_organization_lineage_names": ["SAGE Publishing"], "type": "journal"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "The International Journal of High Performance Computing Applications", "raw_type": "journal-article"}, "cited_by_count": 2, "is_retracted": false}, {"id": "https://openalex.org/W4411358739", "doi": "https://doi.org/10.1109/mnet.2025.3580618", "title": "6G Enabled Generative AI Services for Secure High-Speed Railway Networks", "publication_year": 2025, "publication_date": "2025-06-17", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5079076598", "display_name": "Ruibin Guo", "orcid": "https://orcid.org/0009-0002-9397-9854", "observed_orcids": ["https://orcid.org/0009-0002-9397-9854"]}, "institutions": [{"id": "https://openalex.org/I21193070", "display_name": "Beijing Jiaotong University", "ror": "https://ror.org/01yj56c84", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I21193070"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Ruibin Guo", "raw_affiliation_strings": ["School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China"], "raw_orcid": "https://orcid.org/0009-0002-9397-9854", "affiliations": [{"raw_affiliation_string": "School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China", "institution_ids": ["https://openalex.org/I21193070"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5076675337", "display_name": "Weiting Zhang", "orcid": "https://orcid.org/0000-0003-2626-4883", "observed_orcids": ["https://orcid.org/0000-0003-2626-4883"]}, "institutions": [{"id": "https://openalex.org/I21193070", "display_name": "Beijing Jiaotong University", "ror": "https://ror.org/01yj56c84", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I21193070"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Weiting Zhang", "raw_affiliation_strings": ["School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China"], "raw_orcid": "https://orcid.org/0000-0003-2626-4883", "affiliations": [{"raw_affiliation_string": "School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China", "institution_ids": ["https://openalex.org/I21193070"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5071654735", "display_name": "Dong Yang", "orcid": "https://orcid.org/0000-0003-3402-3668", "observed_orcids": ["https://orcid.org/0000-0003-3402-3668"]}, "institutions": [{"id": "https://openalex.org/I21193070", "display_name": "Beijing Jiaotong University", "ror": "https://ror.org/01yj56c84", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I21193070"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Dong Yang", "raw_affiliation_strings": ["School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China"], "raw_orcid": "https://orcid.org/0000-0003-3402-3668", "affiliations": [{"raw_affiliation_string": "School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China", "institution_ids": ["https://openalex.org/I21193070"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100648099", "display_name": "Hongchao Wang", "orcid": "https://orcid.org/0000-0001-7905-2513", "observed_orcids": ["https://orcid.org/0000-0001-7905-2513"]}, "institutions": [{"id": "https://openalex.org/I21193070", "display_name": "Beijing Jiaotong University", "ror": "https://ror.org/01yj56c84", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I21193070"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Hongchao Wang", "raw_affiliation_strings": ["School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China"], "raw_orcid": "https://orcid.org/0000-0001-7905-2513", "affiliations": [{"raw_affiliation_string": "School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China", "institution_ids": ["https://openalex.org/I21193070"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5091713195", "display_name": "Hongke Zhang", "orcid": "https://orcid.org/0000-0001-8906-813X", "observed_orcids": ["https://orcid.org/0000-0001-8906-813X"]}, "institutions": [{"id": "https://openalex.org/I21193070", "display_name": "Beijing Jiaotong University", "ror": "https://ror.org/01yj56c84", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I21193070"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Hongke Zhang", "raw_affiliation_strings": ["School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China"], "raw_orcid": "https://orcid.org/0000-0001-8906-813X", "affiliations": [{"raw_affiliation_string": "School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China", "institution_ids": ["https://openalex.org/I21193070"]}]}], "abstract_inverted_index": {"Running": [0], "trains": [1], "at": [2], "higher": [3], "speeds": [4], "is": [5, 175], "a": [6, 43, 135, 139, 164], "crucial": [7], "trend": [8], "in": [9, 119, 129], "railway": [10, 36, 49], "development.": [11], "For": [12], "future": [13], "ultra-high-speed": [14], "scenarios,": [15], "the": [16, 80, 120, 160, 196, 207], "sixth-generation": [17], "mobile": [18], "communication": [19, 186], "technology": [20], "(6G)": [21], "can": [22, 103], "provide": [23], "reliable": [24], "transmission": [25, 123], "and": [26, 66, 73, 87, 94, 100, 114, 153, 169, 177], "enable": [27], "generative": [28], "artificial": [29], "intelligence": [30], "(GAI)": [31], "services": [32, 174, 211], "for": [33, 48, 97, 138, 172], "secure": [34], "high-speed": [35], "networks.": [37, 131], "In": [38], "this": [39], "paper,": [40], "we": [41, 107, 133], "propose": [42, 108], "6G-enabled": [44], "GAI": [45, 76, 90, 173, 210], "service": [46], "architecture": [47, 57], "networks": [50], "to": [51, 70, 124, 142, 184], "enhance": [52], "train": [53], "operation": [54], "safety.": [55], "This": [56], "leverages": [58], "integrated": [59, 122], "air-ground-train": [60, 121], "transmission,": [61], "unmanned": [62], "aerial": [63], "vehicle": [64], "(UAV),": [65], "distributed": [67, 167], "learning": [68], "mechanisms": [69], "facilitate": [71], "training": [72, 168], "inference": [74, 170], "of": [75, 84, 151, 198, 209], "services.": [77], "By": [78], "analyzing": [79], "massive": [81], "real-time": [82], "data": [83], "critical": [85], "equipment": [86, 98], "areas": [88], "with": [89], "techniques,": [91], "early": [92], "warnings": [93], "customized": [95], "guidance": [96], "breakdowns": [99], "natural": [101], "disasters": [102], "be": [104], "provided.": [105], "Specifically,": [106], "integrating": [109], "time-sensitive": [110], "networking": [111], "into": [112], "6G": [113], "jointly": [115], "allocating": [116], "time-spectrum": [117], "resources": [118], "guarantee": [125], "deterministic": [126], "end-to-end": [127], "latency": [128], "heterogeneous": [130], "Secondly,": [132], "develop": [134], "self-organization": [136, 161], "mechanism": [137], "UAV": [140], "cluster": [141, 147], "perform": [143], "various": [144], "functions.": [145], "The": [146], "comprises": [148], "different": [149], "types": [150], "UAVs": [152], "multiple": [154], "intelligent": [155], "algorithms": [156], "are": [157, 182, 212], "utilized": [158], "during": [159], "process.": [162], "Furthermore,": [163], "federated": [165], "learning-based": [166, 192], "framework": [171], "studied,": [176], "several": [178], "parallel": [179], "computing": [180], "techniques": [181], "proposed": [183], "reduce": [185], "overhead.": [187], "Simulation": [188], "results": [189], "demonstrate": [190], "that": [191], "resource": [193], "allocation": [194], "promotes": [195], "timeliness": [197], "model": [199], "inference.": [200], "Open": [201], "research": [202], "issues": [203], "on": [204], "further": [205], "improving": [206], "performance": [208], "also": [213], "studied.": [214]}, "primary_topic": {"id": "https://openalex.org/T11133", "display_name": "UAV Applications and Optimization", "score": 0.2590000033378601, "subfield": {"id": "https://openalex.org/subfields/2202", "display_name": "Aerospace Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T11133", "display_name": "UAV Applications and Optimization", "score": 0.2590000033378601, "subfield": {"id": "https://openalex.org/subfields/2202", "display_name": "Aerospace Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11458", "display_name": "Advanced Wireless Communication Technologies", "score": 0.17419999837875366, "subfield": {"id": "https://openalex.org/subfields/2208", "display_name": "Electrical and Electronic Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.1453000009059906, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/mnet.2025.3580618", "is_oa": false, "landing_page_url": "https://doi.org/10.1109/mnet.2025.3580618", "pdf_url": null, "source": {"id": "https://openalex.org/S186584794", "display_name": "IEEE Network", "issn_l": "0890-8044", "issn": ["0890-8044", "1558-156X"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-3", "norway-1"], "host_organization": "https://openalex.org/P4310319808", "host_organization_name": "Institute of Electrical and Electronics Engineers", "host_organization_lineage": ["https://openalex.org/P4310319808"], "host_organization_lineage_names": ["Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Network", "raw_type": "journal-article"}, "cited_by_count": 1, "is_retracted": false}, {"id": "https://openalex.org/W4417199959", "doi": "https://doi.org/10.1038/s41598-025-32087-7", "title": "A privacy preserving intrusion detection framework for IIoT in 6G networks using homomorphic encryption and graph neural networks", "publication_year": 2025, "publication_date": "2025-12-10", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5104126457", "display_name": "Binjie Hua", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I183006215", "display_name": "Shijiazhuang University", "ror": "https://ror.org/028rmam09", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I183006215"]}], "countries": ["CN"], "is_corresponding": true, "raw_author_name": "Binjie Hua", "raw_affiliation_strings": ["School of Future Information Technology, Shijiazhuang University, Shijiazhuang, 050035, Hebei, China. hbj.king.1974@163.com"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Future Information Technology, Shijiazhuang University, Shijiazhuang, 050035, Hebei, China. hbj.king.1974@163.com", "institution_ids": ["https://openalex.org/I183006215"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5017270732", "display_name": "Haiyan Xi", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Haiyan Xi", "raw_affiliation_strings": ["Hebei Institute of Radio and Television Science and Technology, Shijiazhuang, 050011, Hebei, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Hebei Institute of Radio and Television Science and Technology, Shijiazhuang, 050011, Hebei, China", "institution_ids": []}]}], "abstract_inverted_index": {"The": [0, 57, 150], "integration": [1, 194], "of": [2, 6, 75, 86, 169], "the": [3, 72, 109, 166, 175], "Industrial": [4], "Internet": [5], "Things": [7], "(IIoT)": [8], "with": [9, 130, 184, 195], "sixth-generation": [10], "(6G)": [11], "networks": [12], "offers": [13], "revolutionary": [14], "connectivity": [15], "and": [16, 29, 53, 67, 90, 99, 115, 139, 161, 180, 193], "ultra-low": [17], "latency,": [18], "although": [19], "it": [20, 144], "also": [21], "increases": [22], "cybersecurity": [23], "vulnerabilities": [24], "due": [25], "to": [26, 62, 118, 198], "resource-limited": [27], "devices": [28], "dynamic": [30, 167], "network": [31], "structures.": [32], "This": [33, 172], "research": [34], "presents": [35], "a": [36], "novel": [37], "privacy-preserving": [38], "intrusion": [39], "detection": [40, 131], "system": [41, 59], "(IDS)": [42], "for": [43, 111, 146, 177], "IIoT": [44, 148, 182], "in": [45, 158, 201], "6G": [46], "settings,": [47], "using": [48], "Graph": [49], "Neural": [50], "Networks": [51], "(GNNs)": [52], "Homomorphic": [54], "Encryption": [55], "(HE).": [56], "proposed": [58], "utilizes": [60], "GNNs": [61], "represent": [63], "intricate": [64], "inter-device": [65], "interactions": [66], "temporal": [68], "traffic": [69], "patterns,": [70], "facilitating": [71], "effective": [73], "identification": [74], "advanced": [76], "threats,": [77], "such": [78], "as": [79, 123], "Advanced": [80], "Persistent": [81], "Threats": [82], "(APTs),": [83], "Distributed": [84], "Denial": [85], "Service": [87], "(DDoS)": [88], "attacks,": [89], "Mirai": [91], "botnet": [92], "attacks.": [93], "HE": [94, 153], "enables": [95], "secure": [96], "distributed": [97], "training": [98], "inference": [100], "by": [101], "executing": [102], "computations": [103], "on": [104], "encrypted": [105], "data,": [106], "thereby": [107], "obviating": [108], "necessity": [110], "centralized": [112], "data": [113], "aggregation": [114], "assuring": [116], "adherence": [117], "rigorous": [119], "privacy": [120, 162], "standards": [121], "like": [122], "GDPR.": [124], "Experimental": [125], "results": [126], "indicate": [127], "outstanding": [128], "performance,": [129], "accuracies": [132], "over": [133], "98%,": [134], "low": [135], "false": [136], "positive": [137], "rates,": [138], "little": [140], "computing": [141], "cost,": [142], "rendering": [143], "appropriate": [145], "resource-constrained": [147], "devices.": [149], "GNN": [151], "+": [152], "model": [154], "surpasses": [155], "state-of-the-art": [156], "approaches": [157], "accuracy,": [159], "scalability,": [160], "protection,": [163], "effectively": [164], "tackling": [165], "difficulties": [168], "6G-IIoT": [170], "ecosystems.": [171], "study": [173], "lays": [174], "groundwork": [176], "secure,": [178], "scalable,": [179], "privacy-compliant": [181], "systems,": [183], "future": [185], "efforts": [186], "directed": [187], "towards": [188], "adversarial": [189], "robustness,": [190], "real-time": [191], "implementation,": [192], "federated": [196], "learning": [197], "improve": [199], "resilience": [200], "highly": [202], "interconnected": [203], "industrial": [204], "settings.": [205]}, "primary_topic": {"id": "https://openalex.org/T10400", "display_name": "Network Security and Intrusion Detection", "score": 0.9907000064849854, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10400", "display_name": "Network Security and Intrusion Detection", "score": 0.9907000064849854, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11598", "display_name": "Internet Traffic Analysis and Secure E-voting", "score": 0.002300000051036477, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.0008999999845400453, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1038/s41598-025-32087-7", "is_oa": true, "landing_page_url": "https://doi.org/10.1038/s41598-025-32087-7", "pdf_url": "https://www.nature.com/articles/s41598-025-32087-7_reference.pdf", "source": {"id": "https://openalex.org/S196734849", "display_name": "Scientific Reports", "issn_l": "2045-2322", "issn": ["2045-2322"], "is_oa": true, "is_in_doaj": true, "is_core": true, "listed_in": ["cwts-core", "doaj", "doyens", "jufo-1", "ki-jl-1", "medline", "norway-1"], "host_organization": "https://openalex.org/P4310319908", "host_organization_name": "Nature Portfolio", "host_organization_lineage": ["https://openalex.org/P4310319908", "https://openalex.org/P4310319965"], "host_organization_lineage_names": ["Nature Portfolio", "Springer Nature"], "type": "journal"}, "license": "cc-by-nc-nd", "license_id": "https://openalex.org/licenses/cc-by-nc-nd", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Scientific Reports", "raw_type": "journal-article"}, "cited_by_count": 8, "is_retracted": false}, {"id": "https://openalex.org/W4416771802", "doi": "https://doi.org/10.1109/ojcoms.2025.3633162", "title": "Impact of Optical Communication Link Error on Large-Scale AI Training in Data Centers", "publication_year": 2025, "publication_date": "2025-01-01", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5010092400", "display_name": "Abbas Abolfathimomtaz", "orcid": "https://orcid.org/0000-0002-3838-4287", "observed_orcids": ["https://orcid.org/0000-0002-3838-4287"]}, "institutions": [{"id": "https://openalex.org/I4210115038", "display_name": "Huawei Technologies (Canada)", "ror": "https://ror.org/026venb53", "country_code": "CA", "type": "company", "lineage": ["https://openalex.org/I2250955327", "https://openalex.org/I4210115038"]}], "countries": ["CA"], "is_corresponding": false, "raw_author_name": "Abbas Abolfathimomtaz", "raw_affiliation_strings": ["Advanced Optical Technology Laboratory, Huawei Technologies Canada Company Ltd., Ottawa, ON, Canada"], "raw_orcid": "https://orcid.org/0000-0002-3838-4287", "affiliations": [{"raw_affiliation_string": "Advanced Optical Technology Laboratory, Huawei Technologies Canada Company Ltd., Ottawa, ON, Canada", "institution_ids": ["https://openalex.org/I4210115038"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5052730838", "display_name": "Hamid Ebrahimzad", "orcid": "https://orcid.org/0009-0009-0740-1986", "observed_orcids": ["https://orcid.org/0009-0009-0740-1986"]}, "institutions": [{"id": "https://openalex.org/I4210115038", "display_name": "Huawei Technologies (Canada)", "ror": "https://ror.org/026venb53", "country_code": "CA", "type": "company", "lineage": ["https://openalex.org/I2250955327", "https://openalex.org/I4210115038"]}], "countries": ["CA"], "is_corresponding": false, "raw_author_name": "Hamid Ebrahimzad", "raw_affiliation_strings": ["Advanced Optical Technology Laboratory, Huawei Technologies Canada Company Ltd., Ottawa, ON, Canada"], "raw_orcid": "https://orcid.org/0009-0009-0740-1986", "affiliations": [{"raw_affiliation_string": "Advanced Optical Technology Laboratory, Huawei Technologies Canada Company Ltd., Ottawa, ON, Canada", "institution_ids": ["https://openalex.org/I4210115038"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5004761966", "display_name": "Chuandong Li", "orcid": "https://orcid.org/0000-0001-6155-4849", "observed_orcids": ["https://orcid.org/0000-0001-6155-4849"]}, "institutions": [{"id": "https://openalex.org/I4210115038", "display_name": "Huawei Technologies (Canada)", "ror": "https://ror.org/026venb53", "country_code": "CA", "type": "company", "lineage": ["https://openalex.org/I2250955327", "https://openalex.org/I4210115038"]}], "countries": ["CA"], "is_corresponding": false, "raw_author_name": "Chuandong Li", "raw_affiliation_strings": ["Advanced Optical Technology Laboratory, Huawei Technologies Canada Company Ltd., Ottawa, ON, Canada"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Advanced Optical Technology Laboratory, Huawei Technologies Canada Company Ltd., Ottawa, ON, Canada", "institution_ids": ["https://openalex.org/I4210115038"]}]}], "abstract_inverted_index": {"Recently,": [0], "data": [1, 24, 37, 75, 168], "centers": [2], "(DCs)": [3], "have": [4], "been": [5], "increasingly": [6], "dedicated": [7], "to": [8, 123, 166], "artificial": [9], "intelligence": [10], "(AI)": [11], "training": [12, 80, 181], "processes.": [13], "To": [14], "enable": [15, 167], "large-scale": [16], "model": [17, 117], "training,": [18], "parallelization": [19], "schemes": [20], "such": [21], "as": [22], "distributed": [23], "parallelism": [25, 29], "(DDP)": [26], "and": [27, 48, 103, 196, 204], "pipeline": [28], "(PP)": [30], "are": [31, 54, 66], "essential.": [32], "Both": [33], "techniques": [34], "require": [35], "extensive": [36], "transmission": [38, 76, 169], "through": [39, 178], "optical": [40, 64, 144, 156], "communication": [41, 93, 215], "links": [42, 53, 65], "within": [43], "a": [44, 150, 214, 218, 234], "DC.": [45], "The": [46, 190], "latency": [47], "power": [49], "consumption": [50], "of": [51, 74, 92, 186], "these": [52], "critical": [55], "factors": [56], "affecting": [57], "DC": [58], "efficiency": [59], "for": [60, 69, 142, 155], "AI": [61, 79], "training.": [62], "Although": [63], "typically": [67], "engineered": [68], "near-error-free": [70], "transmission,": [71], "the": [72, 90, 97, 101, 116, 136, 162, 182, 198, 230], "impact": [73], "errors": [77, 95, 111], "during": [78], "remains": [81], "insufficiently": [82], "explored.": [83], "In": [84], "this": [85], "work,": [86], "we": [87, 148], "analytically": [88], "investigate": [89], "effect": [91], "link": [94, 110, 130, 145, 202, 216, 235], "on": [96, 226], "learning": [98], "process": [99], "under": [100], "DDP": [102, 227], "PP": [104, 231], "schemes.": [105], "Our": [106], "analysis": [107], "reveals": [108], "that": [109, 213], "introduce": [112], "bounded": [113, 205], "noise": [114], "into": [115], "weights,": [118], "allowing": [119], "weight": [120, 206], "error": [121, 132], "levels": [122], "be": [124], "controlled": [125], "by": [126, 180], "maintaining": [127], "an": [128], "appropriate": [129], "bit": [131], "rate": [133], "(BER).": [134], "Relaxing": [135], "error-free": [137], "requirement": [138], "opens": [139], "new": [140], "opportunities": [141], "optimizing": [143], "performance.": [146], "Specifically,": [147], "propose": [149], "novel": [151], "protocol": [152], "stack": [153], "layer": [154], "links,": [157], "with": [158, 170, 217], "minimal": [159], "deviation": [160], "from": [161], "IEEE": [163], "802.3bs": [164], "standard,": [165], "reduced": [171], "latency.": [172], "We": [173], "validate": [174], "our": [175, 210], "theoretical": [176, 199], "findings": [177], "simulations": [179, 211], "ChatGPT2": [183], "model,": [184], "consisting": [185], "124": [187], "million": [188], "parameters.": [189], "results": [191], "highlight": [192], "several": [193], "practical": [194], "implications": [195], "confirm": [197], "relationship": [200], "between": [201], "BER": [203, 219, 236], "noise.": [207], "For": [208], "example,": [209], "demonstrate": [212], "less": [220, 237], "than": [221, 238], "1e-4": [222], "has": [223], "negligible": [224], "impacts": [225], "performance,": [228], "while": [229], "method": [232], "requires": [233], "1e-5.": [239]}, "primary_topic": {"id": "https://openalex.org/T10847", "display_name": "Advanced Optical Network Technologies", "score": 0.31130000948905945, "subfield": {"id": "https://openalex.org/subfields/2208", "display_name": "Electrical and Electronic Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10847", "display_name": "Advanced Optical Network Technologies", "score": 0.31130000948905945, "subfield": {"id": "https://openalex.org/subfields/2208", "display_name": "Electrical and Electronic Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10232", "display_name": "Optical Network Technologies", "score": 0.07940000295639038, "subfield": {"id": "https://openalex.org/subfields/2208", "display_name": "Electrical and Electronic Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10138", "display_name": "Network Traffic and Congestion Control", "score": 0.0575999990105629, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/ojcoms.2025.3633162", "is_oa": true, "landing_page_url": "https://doi.org/10.1109/ojcoms.2025.3633162", "pdf_url": null, "source": {"id": "https://openalex.org/S4210202420", "display_name": "IEEE Open Journal of the Communications Society", "issn_l": "2644-125X", "issn": ["2644-125X"], "is_oa": true, "is_in_doaj": true, "is_core": true, "listed_in": ["cwts-core", "doaj", "jufo-1", "norway-1"], "host_organization": "https://openalex.org/P4310316002", "host_organization_name": "IEEE Communications Society", "host_organization_lineage": ["https://openalex.org/P4310316002", "https://openalex.org/P4310319808"], "host_organization_lineage_names": ["IEEE Communications Society", "Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Open Journal of the Communications Society", "raw_type": "journal-article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4417003575", "doi": "https://doi.org/10.1109/mcom.001.2500165", "title": "SplitFrozen: Split Learning with Device-Side Model Frozen for Fine-Tuning LLM on Heterogeneous Resource-Constrained Devices", "publication_year": 2025, "publication_date": "2025-12-04", "type": "article", "language": null, "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Jian Ma", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I139759216", "display_name": "Beijing University of Posts and Telecommunications", "ror": "https://ror.org/04w9fbh59", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I139759216"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Jian Ma", "raw_affiliation_strings": ["Beijing University of Posts and Telecommunications,China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Beijing University of Posts and Telecommunications,China", "institution_ids": ["https://openalex.org/I139759216"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5090327058", "display_name": "Xinchen Lyu", "orcid": "https://orcid.org/0000-0003-0404-4310", "observed_orcids": ["https://orcid.org/0000-0003-0404-4310"]}, "institutions": [{"id": "https://openalex.org/I139759216", "display_name": "Beijing University of Posts and Telecommunications", "ror": "https://ror.org/04w9fbh59", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I139759216"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Xinchen Lyu", "raw_affiliation_strings": ["Beijing University of Posts and Telecommunications,China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Beijing University of Posts and Telecommunications,China", "institution_ids": ["https://openalex.org/I139759216"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5050870535", "display_name": "Jun Jiang", "orcid": "https://orcid.org/0000-0002-4558-4930", "observed_orcids": ["https://orcid.org/0000-0002-4558-4930"]}, "institutions": [{"id": "https://openalex.org/I4210136793", "display_name": "Peng Cheng Laboratory", "ror": "https://ror.org/03qdqbt06", "country_code": "CN", "type": "facility", "lineage": ["https://openalex.org/I4210136793"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Jun Jiang", "raw_affiliation_strings": ["Pengcheng Laboratory,China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Pengcheng Laboratory,China", "institution_ids": ["https://openalex.org/I4210136793"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5018587643", "display_name": "Qimei Cui", "orcid": "https://orcid.org/0000-0003-1720-220X", "observed_orcids": ["https://orcid.org/0000-0003-1720-220X"]}, "institutions": [{"id": "https://openalex.org/I139759216", "display_name": "Beijing University of Posts and Telecommunications", "ror": "https://ror.org/04w9fbh59", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I139759216"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Qimei Cui", "raw_affiliation_strings": ["Beijing University of Posts and Telecommunications,China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Beijing University of Posts and Telecommunications,China", "institution_ids": ["https://openalex.org/I139759216"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100666106", "display_name": "Haipeng Yao", "orcid": "https://orcid.org/0000-0003-1391-7363", "observed_orcids": ["https://orcid.org/0000-0003-1391-7363"]}, "institutions": [{"id": "https://openalex.org/I139759216", "display_name": "Beijing University of Posts and Telecommunications", "ror": "https://ror.org/04w9fbh59", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I139759216"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Haipeng Yao", "raw_affiliation_strings": ["Beijing University of Posts and Telecommunications,China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Beijing University of Posts and Telecommunications,China", "institution_ids": ["https://openalex.org/I139759216"]}]}, {"author_position": "last", "author": {"id": null, "display_name": "Xiaofeng Tao", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I139759216", "display_name": "Beijing University of Posts and Telecommunications", "ror": "https://ror.org/04w9fbh59", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I139759216"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Xiaofeng Tao", "raw_affiliation_strings": ["Beijing University of Posts and Telecommunications,China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Beijing University of Posts and Telecommunications,China", "institution_ids": ["https://openalex.org/I139759216"]}]}], "abstract_inverted_index": {"Fine-tuning": [0], "large": [1], "language": [2], "models": [3], "(LLMs)": [4], "on": [5, 18, 57, 127, 170, 177], "private,": [6], "on-device": [7], "data": [8, 32], "can": [9, 112], "empower": [10], "tailored": [11], "personalized": [12], "AI": [13], "agents.": [14], "However,": [15], "fine-tuning": [16, 46, 56, 70], "LLMs": [17, 63], "resource-constrained": [19, 74], "edge": [20], "devices": [21, 75], "faces": [22], "significant": [23], "challenges,": [24], "including": [25], "excessive": [26], "computation": [27], "overhead,": [28], "device": [29], "heterogeneity,": [30], "and": [31, 68, 106, 117, 123, 133, 141, 158], "imbalance.": [33], "This": [34], "article": [35], "proposes": [36], "SplitFrozen,": [37], "a": [38], "split": [39], "learning": [40], "framework": [41, 61], "that": [42, 137], "enables": [43], "efficient": [44], "LLM": [45], "by": [47, 102, 143], "strategically": [48], "freezing": [49], "deviceside": [50], "model": [51, 145, 176], "layers": [52, 67], "while": [53, 151], "centralizing": [54], "parameter-efficient": [55], "the": [58, 91, 130, 166], "server.": [59], "Our": [60], "partitions": [62], "into": [64, 90], "device-side": [65, 156], "frozen": [66], "server-side": [69, 82, 92], "layers,": [71], "where": [72], "heterogeneous": [73], "execute": [76], "only": [77], "forward": [78], "propagation.": [79, 110], "To": [80], "minimize": [81], "training": [83, 100, 161], "costs,": [84], "we": [85], "integrate": [86], "Low-Rank": [87], "Adaptation": [88], "(LoRA)": [89], "layers.": [93], "A": [94], "pipeline": [95], "parallelism": [96], "strategy": [97], "further": [98], "optimizes": [99], "efficiency": [101], "decoupling": [103], "device-server": [104], "computations": [105, 157], "leveraging": [107], "decomposed": [108], "backward": [109], "SplitFrozen": [111, 138, 169], "also": [113, 164], "implement": [114], "differential": [115], "privacy": [116, 121], "compression": [118], "techniques": [119], "for": [120], "preservation": [122], "transmission": [124], "reduction.": [125], "Experiments": [126, 163], "GPT-2": [128], "with": [129], "MRPC,": [131], "MNLI-matched,": [132], "SST-2": [134], "datasets": [135], "demonstrate": [136], "outperforms": [139], "FedLoRA": [140], "SplitLoRA": [142], "69.4%": [144], "accuracy": [146], "under": [147], "extremely": [148], "imbalanced": [149], "data,": [150], "reducing": [152], "up": [153], "to": [154], "86.8%": [155], "50.2%": [159], "total": [160], "time.": [162], "validate": [165], "scalability": [167], "of": [168], "content": [171], "generation": [172], "task": [173], "using": [174], "Llama-3.2": [175], "GSM8K": [178], "dataset.": [179]}, "primary_topic": {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.9883000254631042, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.9883000254631042, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.00419999985024333, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11307", "display_name": "Domain Adaptation and Few-Shot Learning", "score": 0.0038999998942017555, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/mcom.001.2500165", "is_oa": false, "landing_page_url": "https://doi.org/10.1109/mcom.001.2500165", "pdf_url": null, "source": {"id": "https://openalex.org/S158797327", "display_name": "IEEE Communications Magazine", "issn_l": "0163-6804", "issn": ["0163-6804", "1558-1896"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-2", "norway-2"], "host_organization": "https://openalex.org/P4310319808", "host_organization_name": "Institute of Electrical and Electronics Engineers", "host_organization_lineage": ["https://openalex.org/P4310319808"], "host_organization_lineage_names": ["Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Communications Magazine", "raw_type": "journal-article"}, "cited_by_count": 4, "is_retracted": false}, {"id": "https://openalex.org/W4415343879", "doi": null, "title": "DistZO2: High-Throughput and Memory-Efficient Zeroth-Order Fine-tuning LLMs with Distributed Parallel Computing", "publication_year": 2025, "publication_date": "2025-07-03", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5025376724", "display_name": "Liangyu Wang", "orcid": "https://orcid.org/0000-0001-5264-1502", "observed_orcids": ["https://orcid.org/0000-0001-5264-1502"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Wang, Liangyu", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": null, "display_name": "Xie, Huanyi", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Xie, Huanyi", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5012651707", "display_name": "Di Wang", "orcid": "https://orcid.org/0000-0003-2964-6473", "observed_orcids": ["https://orcid.org/0000-0003-2964-6473"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Wang, Di", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Fine-tuning": [0], "large": [1], "language": [2], "models": [3, 28], "(LLMs)": [4], "remains": [5, 66], "resource-intensive": [6], "due": [7], "to": [8, 26, 48, 116, 169], "their": [9], "sheer": [10], "scale.": [11], "While": [12], "zeroth-order": [13, 88, 167], "(ZO)": [14], "optimization": [15], "provides": [16], "a": [17, 61, 82, 125, 146, 189], "memory-efficient": [18, 84], "alternative": [19], "by": [20, 31, 44, 68, 141], "eliminating": [21], "backward": [22], "passes,": [23], "its": [24, 69], "application": [25], "multi-hundred-billion-parameter": [27], "is": [29], "constrained": [30], "GPU": [32], "memory": [33, 42, 50, 175], "and": [34, 51, 72, 123, 133, 154], "compute": [35], "throughput.": [36, 75, 181], "The": [37], "ZO2": [38, 65, 193], "framework": [39, 85], "addresses": [40], "the": [41, 103, 117], "bottleneck": [43], "offloading": [45], "model": [46], "parameters": [47], "CPU": [49], "overlapping": [52], "transformer": [53], "block": [54], "transfer": [55], "with": [56, 194], "dual": [57], "forward": [58, 106], "computation": [59], "on": [60, 185], "single": [62], "GPU.": [63], "However,": [64], "limited": [67], "single-device": [70], "execution": [71], "achieves": [73, 188], "modest": [74], "In": [76, 182], "this": [77], "work,": [78], "we": [79, 144], "present": [80], "DistZO2,": [81], "high-throughput,": [83], "for": [86], "distributed": [87, 195], "fine-tuning": [89, 168], "of": [90, 120], "LLMs.": [91], "DistZO2": [92, 165, 187], "introduces": [93], "three": [94], "parallel": [95], "strategies:": [96], "(1)": [97], "Perturbation": [98], "Parallelism": [99, 113, 128], "(PertP),": [100], "which": [101], "parallelizes": [102], "two": [104], "perturbed": [105], "passes": [107], "across": [108, 157], "devices;": [109], "(2)": [110], "Distributed": [111], "Data": [112], "(DDP),": [114], "adapted": [115], "scalar-gradient": [118], "nature": [119], "ZO": [121], "training;": [122], "(3)": [124], "unified": [126], "2D": [127], "design": [129], "that": [130, 150], "combines": [131], "PertP": [132], "DDP.": [134], "To": [135], "further": [136], "mitigate": [137], "communication": [138, 148], "bottlenecks": [139], "introduced": [140], "parameter": [142, 152], "offloading,": [143], "propose": [145], "hardware-aware": [147], "strategy": [149], "slices": [151], "blocks": [153], "redistributes": [155], "them": [156], "GPUs": [158], "via": [159], "high-speed": [160], "interconnects": [161], "such": [162], "as": [163], "NVLink.": [164], "scales": [166], "modern": [170], "multi-GPU": [171], "systems,": [172], "preserving": [173], "ZO2's": [174], "efficiency": [176], "while": [177], "substantially": [178], "improving": [179], "training": [180], "our": [183], "experiments": [184], "OPT-175B,": [186], "3x": [190], "speedup": [191], "over": [192], "computing.": [196], "DistZO2's": [197], "code": [198], "has": [199], "been": [200], "open-sourced": [201], "in": [202], "https://github.com/liangyuwang/zo2.": [203]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.6039999723434448, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.6039999723434448, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.1907999962568283, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.07750000059604645, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2507.03211", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2507.03211", "pdf_url": "https://arxiv.org/pdf/2507.03211", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4412939970", "doi": "https://doi.org/10.1109/tc.2025.3595078", "title": "Parallelization Strategies for DeepMD-Kit Using OpenMP: Enhancing Efficiency in Machine Learning-Based Molecular Simulations", "publication_year": 2025, "publication_date": "2025-08-04", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Qi Du", "orcid": "https://orcid.org/0009-0008-1584-5360", "observed_orcids": ["https://orcid.org/0009-0008-1584-5360"]}, "institutions": [{"id": "https://openalex.org/I16609230", "display_name": "Hunan University", "ror": "https://ror.org/05htk5m33", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I16609230"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Qi Du", "raw_affiliation_strings": ["College of Computer Science and Electronic Engineering, Hunan University, Changsha, China", "College of Computer Science and Electronic Engineering at Hunan University, Changsha, China"], "raw_orcid": "https://orcid.org/0009-0008-1584-5360", "affiliations": [{"raw_affiliation_string": "College of Computer Science and Electronic Engineering, Hunan University, Changsha, China", "institution_ids": ["https://openalex.org/I16609230"]}, {"raw_affiliation_string": "College of Computer Science and Electronic Engineering at Hunan University, Changsha, China", "institution_ids": ["https://openalex.org/I16609230"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Feng Wang", "orcid": "https://orcid.org/0009-0002-4009-6783", "observed_orcids": ["https://orcid.org/0009-0002-4009-6783"]}, "institutions": [{"id": "https://openalex.org/I16609230", "display_name": "Hunan University", "ror": "https://ror.org/05htk5m33", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I16609230"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Feng Wang", "raw_affiliation_strings": ["College of Computer Science and Electronic Engineering, Hunan University, Changsha, China", "College of Computer Science and Electronic Engineering at Hunan University, Changsha, China"], "raw_orcid": "https://orcid.org/0009-0002-4009-6783", "affiliations": [{"raw_affiliation_string": "College of Computer Science and Electronic Engineering, Hunan University, Changsha, China", "institution_ids": ["https://openalex.org/I16609230"]}, {"raw_affiliation_string": "College of Computer Science and Electronic Engineering at Hunan University, Changsha, China", "institution_ids": ["https://openalex.org/I16609230"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5054335730", "display_name": "Chengkun Wu", "orcid": "https://orcid.org/0000-0002-9688-5311", "observed_orcids": ["https://orcid.org/0000-0002-9688-5311"]}, "institutions": [{"id": "https://openalex.org/I170215575", "display_name": "National University of Defense Technology", "ror": "https://ror.org/05d2yfz11", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I170215575"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Chengkun Wu", "raw_affiliation_strings": ["College of Computer Science and Technology, National University of Defense Technology, Changsha, China", "College of Computer Science and Technology at National University of Defense Technology, Changsha, China"], "raw_orcid": "https://orcid.org/0000-0002-9688-5311", "affiliations": [{"raw_affiliation_string": "College of Computer Science and Technology, National University of Defense Technology, Changsha, China", "institution_ids": ["https://openalex.org/I170215575"]}, {"raw_affiliation_string": "College of Computer Science and Technology at National University of Defense Technology, Changsha, China", "institution_ids": ["https://openalex.org/I170215575"]}]}], "abstract_inverted_index": {"DeepMD-kit": [0, 22], "enables": [1], "deep": [2], "learning-based": [3], "molecular": [4], "dynamics": [5], "(MD)": [6], "simulations": [7], "that": [8], "require": [9], "efficient": [10], "parallelization": [11], "to": [12, 27, 57, 59, 69, 97], "leverage": [13], "modern": [14], "HPC": [15], "architectures.": [16], "In": [17], "this": [18], "work,": [19], "we": [20], "optimize": [21], "using": [23], "advanced": [24], "OpenMP": [25], "strategies": [26], "improve": [28], "scalability": [29], "and": [30, 53, 62, 87], "computational": [31], "efficiency": [32, 67], "on": [33], "an": [34], "ARMv8": [35], "processor-based": [36], "server.": [37], "Our": [38], "optimizations": [39], "include": [40], "data": [41], "parallelism": [42], "for": [43], "neural": [44], "network": [45], "inference,": [46], "force": [47], "calculation": [48], "acceleration,": [49], "NUMAaware": [50], "memory": [51, 65], "management,": [52], "synchronization": [54, 99], "reductions,": [55], "leading": [56], "up": [58], "4.1\u00d7": [60], "speedup": [61, 78], "82%": [63], "higher": [64], "band-width": [66], "compared": [68], "the": [70], "baseline": [71], "implementation.": [72], "Strong": [73], "scaling": [74], "analysis": [75], "demonstrates": [76], "superlinear": [77], "at": [79, 93], "mid-range": [80], "core": [81], "counts,": [82], "with": [83], "improved": [84], "workload": [85], "balancing": [86], "vectorized": [88], "computations.": [89], "However,": [90], "challenges": [91], "remain": [92], "ultra-large": [94], "scales": [95], "due": [96], "increasing": [98], "overhead.": [100]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 1.0, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 1.0, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10002", "display_name": "Advanced Chemical Physics Studies", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/3107", "display_name": "Atomic and Molecular Physics, and Optics"}, "field": {"id": "https://openalex.org/fields/31", "display_name": "Physics and Astronomy"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10481", "display_name": "Computer Graphics and Visualization Techniques", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/1704", "display_name": "Computer Graphics and Computer-Aided Design"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/tc.2025.3595078", "is_oa": false, "landing_page_url": "https://doi.org/10.1109/tc.2025.3595078", "pdf_url": null, "source": {"id": "https://openalex.org/S157670870", "display_name": "IEEE Transactions on Computers", "issn_l": "0018-9340", "issn": ["0018-9340", "0016-9340", "1557-9956", "2326-3814"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-3", "norway-2"], "host_organization": "https://openalex.org/P4310319808", "host_organization_name": "Institute of Electrical and Electronics Engineers", "host_organization_lineage": ["https://openalex.org/P4310319808"], "host_organization_lineage_names": ["Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Transactions on Computers", "raw_type": "journal-article"}, "cited_by_count": 1, "is_retracted": false}, {"id": "https://openalex.org/W4414764129", "doi": "https://doi.org/10.48550/arxiv.2509.05679", "title": "Distributed Deep Learning using Stochastic Gradient Staleness", "publication_year": 2025, "publication_date": "2025-09-06", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5101583696", "display_name": "Viet Hoang Pham", "orcid": "https://orcid.org/0009-0001-7937-8086", "observed_orcids": ["https://orcid.org/0009-0001-7937-8086"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Pham, Viet Hoang", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5003982643", "display_name": "Hyo\u2010Sung Ahn", "orcid": "https://orcid.org/0000-0002-7939-0093", "observed_orcids": ["https://orcid.org/0000-0002-7939-0093"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Ahn, Hyo-Sung", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Despite": [0], "the": [1, 13, 24, 89, 93, 109, 156], "notable": [2], "success": [3], "of": [4, 45, 95], "deep": [5, 39, 71], "neural": [6], "networks": [7], "(DNNs)": [8], "in": [9, 87, 99, 119], "solving": [10], "complex": [11], "tasks,": [12], "training": [14, 51, 62, 96, 120, 125], "process": [15], "still": [16], "remains": [17], "considerable": [18], "challenges.": [19], "A": [20], "primary": [21], "obstacle": [22], "is": [23, 127, 140, 149], "substantial": [25], "time": [26], "required": [27], "for": [28, 69], "training,": [29], "particularly": [30], "as": [31], "high": [32], "performing": [33], "DNNs": [34], "tend": [35], "to": [36, 116, 130, 132, 151], "become": [37], "increasingly": [38], "(characterized": [40], "by": [41], "a": [42, 60], "larger": [43], "number": [44], "hidden": [46], "layers)": [47], "and": [48, 75], "require": [49], "extensive": [50], "datasets.": [52], "To": [53], "address": [54], "these": [55], "challenges,": [56], "this": [57], "paper": [58], "introduces": [59], "distributed": [61, 124], "method": [63, 126], "that": [64], "integrates": [65], "two": [66], "prominent": [67], "strategies": [68], "accelerating": [70], "learning:": [72], "data": [73, 97], "parallelism": [74], "fully": [76], "decoupled": [77], "parallel": [78], "backpropagation": [79, 110], "algorithm.": [80, 111], "By": [81], "utilizing": [82], "multiple": [83], "computational": [84], "units": [85], "operating": [86], "parallel,": [88], "proposed": [90, 123], "approach": [91], "enhances": [92], "amount": [94], "processed": [98], "each": [100], "iteration": [101], "while": [102], "mitigating": [103], "locking": [104], "issues": [105], "commonly": [106], "associated": [107], "with": [108], "These": [112], "features": [113], "collectively": [114], "contribute": [115], "significant": [117], "improvements": [118], "efficiency.": [121], "The": [122], "rigorously": [128], "proven": [129], "converge": [131], "critical": [133], "points": [134], "under": [135], "certain": [136], "conditions.": [137], "Its": [138], "effectiveness": [139], "further": [141], "demonstrated": [142], "through": [143], "empirical": [144], "evaluations,": [145], "wherein": [146], "an": [147], "DNN": [148], "trained": [150], "perform": [152], "classification": [153], "tasks": [154], "on": [155], "CIFAR-10": [157], "dataset.": [158]}, "primary_topic": {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.9897000193595886, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.9897000193595886, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.0027000000700354576, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10320", "display_name": "Neural Networks and Applications", "score": 0.0010000000474974513, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2509.05679", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2509.05679", "pdf_url": "https://arxiv.org/pdf/2509.05679", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4413343110", "doi": "https://doi.org/10.1155/que2/5577635", "title": "LFQAP2: Large\u2010Scale Quantum Artificial Intelligence Training Platform Based on Supercomputing", "publication_year": 2025, "publication_date": "2025-01-01", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5100327518", "display_name": "Xin Zhang", "orcid": "https://orcid.org/0000-0002-7947-916X", "observed_orcids": ["https://orcid.org/0000-0002-7947-916X"]}, "institutions": [{"id": "https://openalex.org/I162868743", "display_name": "Tianjin University", "ror": "https://ror.org/012tb2g32", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I162868743"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Xin Zhang", "raw_affiliation_strings": ["College of Intelligence and Computing ,  Tianjin University ,  Tianjin ,  China , tju.edu.cn"], "raw_orcid": "https://orcid.org/0000-0002-7947-916X", "affiliations": [{"raw_affiliation_string": "College of Intelligence and Computing ,  Tianjin University ,  Tianjin ,  China , tju.edu.cn", "institution_ids": ["https://openalex.org/I162868743"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5116337569", "display_name": "Xiaoyu Li", "orcid": "https://orcid.org/0000-0001-6586-4705", "observed_orcids": ["https://orcid.org/0000-0001-6586-4705"]}, "institutions": [{"id": "https://openalex.org/I150229711", "display_name": "University of Electronic Science and Technology of China", "ror": "https://ror.org/04qr3zq92", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I150229711"]}, {"id": "https://openalex.org/I4210162809", "display_name": "Kashi University", "ror": "https://ror.org/055a4rj94", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I4210162809"]}], "countries": ["CN"], "is_corresponding": true, "raw_author_name": "Xiaoyu Li", "raw_affiliation_strings": ["Institute of Electronics and Information Industry Technology of Kash ,  Kash ,  China", "School of Information and Software Engineering ,  University of Electronic Science and Technology of China ,  Chengdu ,  China , uestc.edu.cn"], "raw_orcid": "https://orcid.org/0000-0001-6586-4705", "affiliations": [{"raw_affiliation_string": "Institute of Electronics and Information Industry Technology of Kash ,  Kash ,  China", "institution_ids": ["https://openalex.org/I4210162809"]}, {"raw_affiliation_string": "School of Information and Software Engineering ,  University of Electronic Science and Technology of China ,  Chengdu ,  China , uestc.edu.cn", "institution_ids": ["https://openalex.org/I150229711"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5066132067", "display_name": "L. F. Wei", "orcid": "https://orcid.org/0000-0003-1533-1550", "observed_orcids": ["https://orcid.org/0000-0003-1533-1550"]}, "institutions": [{"id": "https://openalex.org/I4800084", "display_name": "Southwest Jiaotong University", "ror": "https://ror.org/00hn7w693", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I4800084"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Lianfu Wei", "raw_affiliation_strings": ["College of Information Science and Technology ,  Southwest Jiaotong University ,  Chengdu ,  China , swjtu.edu.cn"], "raw_orcid": "https://orcid.org/0000-0003-1533-1550", "affiliations": [{"raw_affiliation_string": "College of Information Science and Technology ,  Southwest Jiaotong University ,  Chengdu ,  China , swjtu.edu.cn", "institution_ids": ["https://openalex.org/I4800084"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5101642381", "display_name": "Qinsheng Zhu", "orcid": "https://orcid.org/0000-0002-8202-7447", "observed_orcids": ["https://orcid.org/0000-0002-8202-7447"]}, "institutions": [{"id": "https://openalex.org/I150229711", "display_name": "University of Electronic Science and Technology of China", "ror": "https://ror.org/04qr3zq92", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I150229711"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Qinsheng Zhu", "raw_affiliation_strings": ["School of Computer Science and Engineering ,  University of Electronic Science and Technology of China ,  Chengdu ,  China , uestc.edu.cn"], "raw_orcid": "https://orcid.org/0000-0002-8202-7447", "affiliations": [{"raw_affiliation_string": "School of Computer Science and Engineering ,  University of Electronic Science and Technology of China ,  Chengdu ,  China , uestc.edu.cn", "institution_ids": ["https://openalex.org/I150229711"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5037715125", "display_name": "Geng Chen", "orcid": "https://orcid.org/0000-0001-5546-0205", "observed_orcids": ["https://orcid.org/0000-0001-5546-0205"]}, "institutions": [{"id": "https://openalex.org/I150229711", "display_name": "University of Electronic Science and Technology of China", "ror": "https://ror.org/04qr3zq92", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I150229711"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Geng Chen", "raw_affiliation_strings": ["School of Computer Science and Engineering ,  University of Electronic Science and Technology of China ,  Chengdu ,  China , uestc.edu.cn"], "raw_orcid": "https://orcid.org/0000-0001-5546-0205", "affiliations": [{"raw_affiliation_string": "School of Computer Science and Engineering ,  University of Electronic Science and Technology of China ,  Chengdu ,  China , uestc.edu.cn", "institution_ids": ["https://openalex.org/I150229711"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Wenjie Sun", "orcid": "https://orcid.org/0009-0002-3193-0676", "observed_orcids": ["https://orcid.org/0009-0002-3193-0676"]}, "institutions": [{"id": "https://openalex.org/I150229711", "display_name": "University of Electronic Science and Technology of China", "ror": "https://ror.org/04qr3zq92", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I150229711"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Wenjie Sun", "raw_affiliation_strings": ["School of Electronic Science and Engineering ,  University of Electronic Science and Technology of China ,  Chengdu ,  China , uestc.edu.cn"], "raw_orcid": "https://orcid.org/0009-0002-3193-0676", "affiliations": [{"raw_affiliation_string": "School of Electronic Science and Engineering ,  University of Electronic Science and Technology of China ,  Chengdu ,  China , uestc.edu.cn", "institution_ids": ["https://openalex.org/I150229711"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5029744962", "display_name": "Lianhui Yu", "orcid": "https://orcid.org/0009-0007-5079-1110", "observed_orcids": ["https://orcid.org/0009-0007-5079-1110"]}, "institutions": [{"id": "https://openalex.org/I150229711", "display_name": "University of Electronic Science and Technology of China", "ror": "https://ror.org/04qr3zq92", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I150229711"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Lianhui Yu", "raw_affiliation_strings": ["School of Physics ,  University of Electronic Science and Technology of China ,  Chengdu ,  China , uestc.edu.cn"], "raw_orcid": "https://orcid.org/0009-0007-5079-1110", "affiliations": [{"raw_affiliation_string": "School of Physics ,  University of Electronic Science and Technology of China ,  Chengdu ,  China , uestc.edu.cn", "institution_ids": ["https://openalex.org/I150229711"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5100781300", "display_name": "Yuexian Hou", "orcid": "https://orcid.org/0000-0002-3238-493X", "observed_orcids": ["https://orcid.org/0000-0002-3238-493X"]}, "institutions": [{"id": "https://openalex.org/I162868743", "display_name": "Tianjin University", "ror": "https://ror.org/012tb2g32", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I162868743"]}], "countries": ["CN"], "is_corresponding": true, "raw_author_name": "Yuexian Hou", "raw_affiliation_strings": ["College of Intelligence and Computing ,  Tianjin University ,  Tianjin ,  China , tju.edu.cn"], "raw_orcid": "https://orcid.org/0000-0002-3238-493X", "affiliations": [{"raw_affiliation_string": "College of Intelligence and Computing ,  Tianjin University ,  Tianjin ,  China , tju.edu.cn", "institution_ids": ["https://openalex.org/I162868743"]}]}], "abstract_inverted_index": {"Quantum": [0], "artificial": [1], "intelligence": [2], "(QAI)": [3], "has": [4, 21, 100], "made": [5], "significant": [6], "progress": [7], "in": [8], "recent": [9], "years.": [10], "However,": [11], "the": [12, 68, 80, 90, 117, 131, 155, 179, 188, 195, 203, 212, 220, 230], "lack": [13], "of": [14, 92, 119, 124, 165, 232], "platforms": [15], "for": [16, 43, 70, 82], "large\u2010scale": [17, 45, 248], "QAI": [18, 46], "models": [19], "training": [20, 44, 231], "constrained": [22], "research": [23, 246], "on": [24, 140, 162, 247], "related": [25], "algorithms,": [26], "such": [27], "as": [28], "quantum": [29, 249], "large": [30], "language": [31, 174], "models.": [32, 47, 250], "In": [33], "this": [34], "paper,": [35], "we": [36, 158], "introduce": [37], "LFQAP2,": [38], "a": [39, 51, 121, 141, 233, 240], "supercomputing\u2010based": [40], "platform": [41, 49, 243], "designed": [42], "The": [48], "employs": [50], "hybrid": [52], "parallelization": [53, 87], "strategy": [54, 88], "that": [55], "integrates": [56, 89], "data": [57, 63, 94], "parallelism": [58, 95], "and": [59, 96, 105, 108, 138, 172, 219], "model": [60, 75, 97, 137, 235], "parallelism.": [61, 98], "For": [62, 74, 178, 202], "parallelism,": [64, 76], "different": [65, 77], "threads": [66], "compute": [67, 79], "loss": [69], "distinct": [71], "sample": [72], "batches.": [73, 85], "processes": [78], "gradients": [81], "separate": [83], "parameter": [84], "This": [86], "strengths": [91], "both": [93], "It": [99], "high": [101], "node\u2010internal": [102], "computing": [103, 151], "efficiency": [104, 191, 199, 215, 224], "excellent": [106], "scalability,": [107], "can": [109], "reduce": [110], "internode": [111], "communication": [112], "overhead,": [113], "limiting": [114], "it": [115], "to": [116, 244], "exchange": [118], "only": [120], "small": [122], "number": [123], "floating\u2010point": [125], "values.": [126], "LFQAP2": [127], "is": [128, 147, 200, 225], "implemented": [129], "using": [130], "MPI": [132], "+": [133], "OpenMP": [134], "parallel": [135], "programming": [136], "deployed": [139], "16\u2010node": [142, 196, 221], "cluster,": [143], "where": [144], "each": [145], "node": [146], "equipped": [148], "with": [149, 184, 208], "32": [150], "cores.": [152], "To": [153], "assess": [154], "platform\u2019s": [156], "efficiency,": [157], "evaluate": [159], "its": [160], "performance": [161], "two": [163], "types": [164], "tasks:": [166], "image": [167], "recognition": [168], "(MNIST": [169], "dataset": [170], "classification)": [171], "natural": [173], "processing": [175], "(word": [176], "embedding).": [177], "16\u2010qubit": [180], "MNIST": [181], "classification": [182], "task": [183, 207], "128": [185], "trainable": [186, 210], "parameters,": [187, 211], "32\u2010thread": [189, 213], "acceleration": [190, 198, 214, 223], "reaches": [192, 217], "0.852,": [193], "while": [194], "multiprocess": [197, 222], "0.357.": [201], "9\u2010qubit": [204], "word": [205], "embedding": [206], "288": [209], "also": [216], "0.843,": [218], "0.598.": [226], "We": [227], "successfully": [228], "completed": [229], "288\u2010parameter": [234], "within": [236], "23": [237], "min,": [238], "providing": [239], "robust": [241], "computational": [242], "support": [245]}, "primary_topic": {"id": "https://openalex.org/T10682", "display_name": "Quantum Computing Algorithms and Architecture", "score": 0.9926999807357788, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10682", "display_name": "Quantum Computing Algorithms and Architecture", "score": 0.9926999807357788, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.006300000008195639, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T13904", "display_name": "Artificial Intelligence Applications", "score": 9.999999747378752e-05, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1155/que2/5577635", "is_oa": true, "landing_page_url": "https://doi.org/10.1155/que2/5577635", "pdf_url": "https://onlinelibrary.wiley.com/doi/pdfdirect/10.1155/que2/5577635", "source": {"id": "https://openalex.org/S4210208943", "display_name": "Quantum Engineering", "issn_l": "2577-0470", "issn": ["2577-0470"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core"], "host_organization": "https://openalex.org/P4310320595", "host_organization_name": "Wiley", "host_organization_lineage": ["https://openalex.org/P4310320595"], "host_organization_lineage_names": ["Wiley"], "type": "journal"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Quantum Engineering", "raw_type": "journal-article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4409106571", "doi": "https://doi.org/10.3390/electronics14071430", "title": "A Hybrid Federated Learning Framework for Privacy-Preserving Near-Real-Time Intrusion Detection in IoT Environments", "publication_year": 2025, "publication_date": "2025-04-02", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5116883721", "display_name": "Glauco Rampone", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I7882870", "display_name": "University of Glasgow", "ror": "https://ror.org/00vtgdb53", "country_code": "GB", "type": "education", "lineage": ["https://openalex.org/I7882870"]}], "countries": ["GB"], "is_corresponding": false, "raw_author_name": "Glauco Rampone", "raw_affiliation_strings": ["School of Mathematics & Statistics, University of Glasgow, Glasgow G1 1QL, UK"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Mathematics & Statistics, University of Glasgow, Glasgow G1 1QL, UK", "institution_ids": ["https://openalex.org/I7882870"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5116883722", "display_name": "Taras Ivaniv", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I9360294", "display_name": "University of Bologna", "ror": "https://ror.org/01111rn36", "country_code": "IT", "type": "education", "lineage": ["https://openalex.org/I9360294"]}], "countries": ["IT"], "is_corresponding": false, "raw_author_name": "Taras Ivaniv", "raw_affiliation_strings": ["Department Statistical Sciences \u201cPaolo Fortunati\u201d, Alma Mater Studiorum Universit\u00e0 di Bologna, 40138 Bologna, Italy"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department Statistical Sciences \u201cPaolo Fortunati\u201d, Alma Mater Studiorum Universit\u00e0 di Bologna, 40138 Bologna, Italy", "institution_ids": ["https://openalex.org/I9360294"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5075183627", "display_name": "Salvatore Rampone", "orcid": "https://orcid.org/0000-0002-2019-2746", "observed_orcids": ["https://orcid.org/0000-0002-2019-2746"]}, "institutions": [{"id": "https://openalex.org/I16337185", "display_name": "University of Sannio", "ror": "https://ror.org/04vc81p87", "country_code": "IT", "type": "education", "lineage": ["https://openalex.org/I16337185"]}], "countries": ["IT"], "is_corresponding": true, "raw_author_name": "Salvatore Rampone", "raw_affiliation_strings": ["Department Law, Economics, Management and Quantitative Methods (DEMM), University of Sannio, 82100 Benevento, Italy"], "raw_orcid": "https://orcid.org/0000-0002-2019-2746", "affiliations": [{"raw_affiliation_string": "Department Law, Economics, Management and Quantitative Methods (DEMM), University of Sannio, 82100 Benevento, Italy", "institution_ids": ["https://openalex.org/I16337185"]}]}], "abstract_inverted_index": {"The": [0, 168, 242, 283], "proliferation": [1], "of": [2, 4, 18, 140, 183, 201, 222, 233, 303], "Internet": [3], "Things": [5], "(IoT)": [6], "devices": [7, 65, 95], "has": [8], "introduced": [9], "significant": [10, 179], "challenges": [11, 219], "in": [12, 15, 54, 113, 181, 226, 313, 322], "cybersecurity,": [13], "particularly": [14], "the": [16, 36, 111, 138, 191, 199, 218, 231, 247, 301, 317], "realm": [17], "intrusion": [19, 52], "detection.": [20], "While": [21], "effective,": [22], "traditional": [23, 146], "centralized": [24, 147], "machine": [25, 89, 324], "learning": [26, 48, 58, 90, 175, 203, 225, 305, 325], "approaches": [27], "often": [28], "compromise": [29], "data": [30, 39, 106, 163, 184], "privacy": [31, 72, 156, 185], "and": [32, 73, 134, 155, 186, 212, 220, 236, 262], "scalability": [33, 187], "due": [34], "to": [35, 107, 110, 122], "need": [37], "for": [38, 50, 127, 238, 256, 260, 264, 294, 310, 319], "aggregation.": [40], "In": [41], "this": [42, 120, 172], "study,": [43], "we": [44], "propose": [45], "a": [46, 81, 88, 97, 145, 227, 270, 280, 307], "federated": [47, 124, 174, 202, 224, 304], "framework": [49], "near-real-time": [51], "detection": [53], "IoT": [55, 94, 314], "environments.": [56], "Federated": [57], "enables": [59], "decentralized": [60], "model": [61, 91, 121, 143, 240], "training": [62, 129, 153, 273], "across": [63, 165], "multiple": [64, 166], "without": [66], "exchanging": [67], "raw": [68], "data,": [69], "thereby": [70], "preserving": [71], "reducing": [74], "communication": [75], "overhead.": [76], "Our": [77, 158, 298], "approach": [78], "builds": [79], "upon": [80], "previously": [82], "proposed": [83], "hybrid": [84, 173], "model,": [85, 148], "which": [86], "combines": [87], "deployed": [92], "on": [93, 150, 279], "with": [96, 204], "second-level": [98], "cloud-based": [99, 205], "analysis.": [100], "This": [101, 195], "previous": [102, 192, 284], "work": [103], "required": [104], "all": [105, 246], "be": [108], "passed": [109], "cloud": [112], "aggregate": [114], "form,": [115], "limiting": [116], "security.": [117], "We": [118, 136, 216, 267], "extend": [119], "incorporate": [123], "learning,": [125], "allowing": [126], "distributed": [128, 228], "while": [130], "maintaining": [131], "high": [132], "accuracy": [133, 151, 254], "privacy.": [135], "evaluate": [137], "performance": [139], "our": [141], "federated-learning-based": [142], "against": [144], "focusing": [149], "retention,": [152], "efficiency,": [154], "preservation.": [157], "experiments": [159], "utilize": [160], "actual": [161], "attack": [162], "partitioned": [164], "nodes.": [167], "results": [169, 243], "demonstrate": [170], "that": [171, 245], "not": [176], "only": [177], "offers": [178], "advantages": [180], "terms": [182], "but": [188], "also": [189, 197, 268, 290], "retains": [190], "competitive": [193], "accuracy.": [194], "paper": [196], "explores": [198], "integration": [200], "infrastructure,": [206], "leveraging": [207], "platforms": [208], "such": [209], "as": [210, 306], "Databricks": [211], "Google": [213], "Cloud": [214], "Storage.": [215], "discuss": [217], "benefits": [221], "implementing": [223], "environment,": [229], "including": [230], "use": [232], "Apache": [234], "Spark": [235], "MLlib": [237], "scalable": [239], "training.": [241], "show": [244], "algorithms": [248], "used": [249], "maintain": [250], "an": [251], "excellent": [252], "identification": [253], "(98%": [255], "logistic": [257], "R=regression,": [258], "97%": [259], "SVM,": [261], "100%": [263], "Random": [265], "Forest).": [266], "report": [269], "very": [271, 285], "short": [272], "time": [274, 288], "(less": [275], "than": [276], "11": [277], "s": [278, 293], "single": [281], "machine).": [282], "low": [286], "application": [287], "is": [289], "confirmed": [291], "(0.16": [292], "over": [295], "1,697,851": [296], "packets).": [297], "findings": [299], "highlight": [300], "potential": [302], "viable": [308], "solution": [309], "enhancing": [311], "cybersecurity": [312], "ecosystems,": [315], "paving": [316], "way": [318], "further": [320], "research": [321], "privacy-preserving": [323], "techniques.": [326]}, "primary_topic": {"id": "https://openalex.org/T10400", "display_name": "Network Security and Intrusion Detection", "score": 0.5092999935150146, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10400", "display_name": "Network Security and Intrusion Detection", "score": 0.5092999935150146, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.4771000146865845, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T13038", "display_name": "Internet of Things and AI", "score": 0.0035000001080334187, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.3390/electronics14071430", "is_oa": true, "landing_page_url": "https://doi.org/10.3390/electronics14071430", "pdf_url": "https://www.mdpi.com/2079-9292/14/7/1430/pdf?version=1743583601", "source": {"id": "https://openalex.org/S4210202905", "display_name": "Electronics", "issn_l": "2079-9292", "issn": ["2079-9292"], "is_oa": true, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "norway-1"], "host_organization": "https://openalex.org/P4310310987", "host_organization_name": "Multidisciplinary Digital Publishing Institute", "host_organization_lineage": ["https://openalex.org/P4310310987"], "host_organization_lineage_names": ["Multidisciplinary Digital Publishing Institute"], "type": "journal"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Electronics", "raw_type": "journal-article"}, "cited_by_count": 28, "is_retracted": false}, {"id": "https://openalex.org/W4411787131", "doi": "https://doi.org/10.1007/s10766-025-00802-6", "title": "Analysis of Model Parallelism for AI Applications on a 64-core RV64 Server CPU", "publication_year": 2025, "publication_date": "2025-06-30", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5094187870", "display_name": "Giulio Malenza", "orcid": "https://orcid.org/0009-0006-4862-7429", "observed_orcids": ["https://orcid.org/0009-0006-4862-7429"]}, "institutions": [{"id": "https://openalex.org/I55143463", "display_name": "University of Turin", "ror": "https://ror.org/048tbm396", "country_code": "IT", "type": "education", "lineage": ["https://openalex.org/I55143463"]}], "countries": ["IT"], "is_corresponding": true, "raw_author_name": "Giulio Malenza", "raw_affiliation_strings": ["Department of Computer Science, University of Turin, Corso Svizzera 185, Turin, 10149, Italy"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Computer Science, University of Turin, Corso Svizzera 185, Turin, 10149, Italy", "institution_ids": ["https://openalex.org/I55143463"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5051009274", "display_name": "Adriano Marques Garcia", "orcid": "https://orcid.org/0000-0003-4796-773X", "observed_orcids": ["https://orcid.org/0000-0003-4796-773X"]}, "institutions": [{"id": "https://openalex.org/I55143463", "display_name": "University of Turin", "ror": "https://ror.org/048tbm396", "country_code": "IT", "type": "education", "lineage": ["https://openalex.org/I55143463"]}], "countries": ["IT"], "is_corresponding": true, "raw_author_name": "Adriano Marques Garcia", "raw_affiliation_strings": ["Department of Computer Science, University of Turin, Corso Svizzera 185, Turin, 10149, Italy"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Computer Science, University of Turin, Corso Svizzera 185, Turin, 10149, Italy", "institution_ids": ["https://openalex.org/I55143463"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5000838509", "display_name": "Robert Birke", "orcid": "https://orcid.org/0000-0003-1144-3707", "observed_orcids": ["https://orcid.org/0000-0003-1144-3707"]}, "institutions": [{"id": "https://openalex.org/I55143463", "display_name": "University of Turin", "ror": "https://ror.org/048tbm396", "country_code": "IT", "type": "education", "lineage": ["https://openalex.org/I55143463"]}], "countries": ["IT"], "is_corresponding": true, "raw_author_name": "Robert Birke", "raw_affiliation_strings": ["Department of Computer Science, University of Turin, Corso Svizzera 185, Turin, 10149, Italy"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Computer Science, University of Turin, Corso Svizzera 185, Turin, 10149, Italy", "institution_ids": ["https://openalex.org/I55143463"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5043408422", "display_name": "Luca Benini", "orcid": "https://orcid.org/0000-0001-8068-3806", "observed_orcids": ["https://orcid.org/0000-0001-8068-3806"]}, "institutions": [{"id": "https://openalex.org/I9360294", "display_name": "University of Bologna", "ror": "https://ror.org/01111rn36", "country_code": "IT", "type": "education", "lineage": ["https://openalex.org/I9360294"]}], "countries": ["IT"], "is_corresponding": false, "raw_author_name": "Luca Benini", "raw_affiliation_strings": ["Department of Computer Science, University of Bologna, Mura Anteo Zamboni, Bologna, 40126, Italy"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Computer Science, University of Bologna, Mura Anteo Zamboni, Bologna, 40126, Italy", "institution_ids": ["https://openalex.org/I9360294"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5089092418", "display_name": "Marco Aldinucci", "orcid": "https://orcid.org/0000-0001-8788-0829", "observed_orcids": ["https://orcid.org/0000-0001-8788-0829"]}, "institutions": [{"id": "https://openalex.org/I55143463", "display_name": "University of Turin", "ror": "https://ror.org/048tbm396", "country_code": "IT", "type": "education", "lineage": ["https://openalex.org/I55143463"]}], "countries": ["IT"], "is_corresponding": false, "raw_author_name": "Marco Aldinucci", "raw_affiliation_strings": ["Department of Computer Science, University of Turin, Corso Svizzera 185, Turin, 10149, Italy"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Computer Science, University of Turin, Corso Svizzera 185, Turin, 10149, Italy", "institution_ids": ["https://openalex.org/I55143463"]}]}], "abstract_inverted_index": null, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9994999766349792, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9994999766349792, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10904", "display_name": "Embedded Systems Design Techniques", "score": 9.999999747378752e-05, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10715", "display_name": "Distributed and Parallel Computing Systems", "score": 9.999999747378752e-05, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1007/s10766-025-00802-6", "is_oa": false, "landing_page_url": "https://doi.org/10.1007/s10766-025-00802-6", "pdf_url": null, "source": {"id": "https://openalex.org/S148521650", "display_name": "International Journal of Parallel Programming", "issn_l": "0885-7458", "issn": ["0885-7458", "1573-7640"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-2", "norway-1"], "host_organization": "https://openalex.org/P4310319900", "host_organization_name": "Springer Science+Business Media", "host_organization_lineage": ["https://openalex.org/P4310319900", "https://openalex.org/P4310319965"], "host_organization_lineage_names": ["Springer Science+Business Media", "Springer Nature"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "International Journal of Parallel Programming", "raw_type": "journal-article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4414830449", "doi": "https://doi.org/10.48550/arxiv.2504.08242", "title": "Jupiter: Fast and Resource-Efficient Collaborative Inference of Generative LLMs on Edge Devices", "publication_year": 2025, "publication_date": "2025-04-11", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5069427601", "display_name": "Shengyuan Ye", "orcid": "https://orcid.org/0000-0001-8867-0655", "observed_orcids": ["https://orcid.org/0000-0001-8867-0655"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Ye, Shengyuan", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5106355629", "display_name": "Bei Ouyang", "orcid": "https://orcid.org/0009-0001-4408-2832", "observed_orcids": ["https://orcid.org/0009-0001-4408-2832"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Ouyang, Bei", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5055161955", "display_name": "Liekang Zeng", "orcid": "https://orcid.org/0000-0003-4800-8768", "observed_orcids": ["https://orcid.org/0000-0003-4800-8768"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Zeng, Liekang", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5106355630", "display_name": "Tianyi Qian", "orcid": "https://orcid.org/0009-0001-2802-2258", "observed_orcids": ["https://orcid.org/0009-0001-2802-2258"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Qian, Tianyi", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100730785", "display_name": "Xiaowen Chu", "orcid": "https://orcid.org/0000-0001-9745-4372", "observed_orcids": ["https://orcid.org/0000-0001-9745-4372"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Chu, Xiaowen", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5039176528", "display_name": "Jian Tang", "orcid": "https://orcid.org/0000-0003-4418-0114", "observed_orcids": ["https://orcid.org/0000-0003-4418-0114"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Tang, Jian", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5100385709", "display_name": "Chen Xu", "orcid": "https://orcid.org/0000-0002-3070-9358", "observed_orcids": ["https://orcid.org/0000-0002-3070-9358"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Chen, Xu", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Generative": [0], "large": [1], "language": [2], "models": [3], "(LLMs)": [4], "have": [5], "garnered": [6], "significant": [7], "attention": [8], "due": [9], "to": [10, 33, 70, 143, 170, 214], "their": [11], "exceptional": [12], "capabilities": [13], "in": [14, 20, 53], "various": [15, 208], "AI": [16, 122], "tasks.": [17], "Traditionally": [18], "deployed": [19], "cloud": [21], "datacenters,": [22], "LLMs": [23], "are": [24], "now": [25], "increasingly": [26], "moving": [27], "towards": [28], "more": [29], "accessible": [30], "edge": [31, 48, 68, 89, 121, 209], "platforms": [32], "protect": [34], "sensitive": [35], "user": [36], "data": [37], "and": [38, 58, 86, 118, 137, 150, 163], "ensure": [39], "privacy": [40], "preservation.": [41], "The": [42], "limited": [43], "computational": [44], "resources": [45], "of": [46, 75, 88, 147], "individual": [47, 76], "devices,": [49, 77], "however,": [50], "can": [51], "result": [52], "excessively": [54], "prolonged": [55], "inference": [56, 192], "latency": [57, 217], "overwhelmed": [59], "memory": [60], "usage.": [61], "While": [62], "existing": [63], "research": [64], "has": [65], "explored": [66], "collaborative": [67, 120], "computing": [69], "break": [71], "the": [72, 97, 101, 144, 148], "resource": [73, 172], "wall": [74], "these": [78], "solutions": [79], "yet": [80], "suffer": [81], "from": [82], "massive": [83], "communication": [84], "overhead": [85], "under-utilization": [87], "resources.": [90], "Furthermore,": [91], "they": [92], "focus": [93], "exclusively": [94], "on": [95, 197], "optimizing": [96], "prefill": [98, 149, 154], "phase,": [99, 155], "neglecting": [100], "crucial": [102], "autoregressive": [103], "decoding": [104, 151, 183], "phase": [105], "for": [106, 124], "generative": [107, 125], "LLMs.": [108], "To": [109], "address": [110], "that,": [111], "we": [112], "propose": [113], "Jupiter,": [114], "a": [115, 130, 135, 158, 165], "fast,": [116], "scalable,": [117], "resource-efficient": [119], "system": [123, 140], "LLM": [126], "inference.": [127], "Jupiter": [128, 156, 176, 202], "introduces": [129], "flexible": [131], "pipelined": [132], "architecture": [133], "as": [134], "principle": [136], "differentiates": [138], "its": [139], "design": [141], "according": [142], "differentiated": [145], "characteristics": [146], "phases.": [152], "For": [153, 174], "submits": [157], "novel": [159], "intra-sequence": [160], "pipeline": [161, 181], "parallelism": [162, 167], "develops": [164], "meticulous": [166], "planning": [168], "strategy": [169], "maximize": [171], "efficiency;": [173], "decoding,": [175, 188], "devises": [177], "an": [178], "effective": [179], "outline-based": [180], "parallel": [182], "mechanism": [184], "combined": [185], "with": [186], "speculative": [187], "which": [189], "further": [190], "magnifies": [191], "acceleration.": [193], "Extensive": [194], "evaluation": [195], "based": [196], "realistic": [198], "implementation": [199], "demonstrates": [200], "that": [201], "remarkably": [203], "outperforms": [204], "state-of-the-art": [205], "approaches": [206], "under": [207], "environment": [210], "setups,": [211], "achieving": [212], "up": [213], "26.1x": [215], "end-to-end": [216], "reduction": [218], "while": [219], "rendering": [220], "on-par": [221], "generation": [222], "quality.": [223]}, "primary_topic": {"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.9624000191688538, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.9624000191688538, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.0071000000461936, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10181", "display_name": "Natural Language Processing Techniques", "score": 0.005400000140070915, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2504.08242", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2504.08242", "pdf_url": "https://arxiv.org/pdf/2504.08242", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4414064990", "doi": "https://doi.org/10.55248/gengpi.6.0825.3115", "title": "Transformers on encrypted federated datasets anchored by blockchain zero-knowledge proofs for privacy-preserving multilingual healthcare diagnostics and equity", "publication_year": 2025, "publication_date": "2025-08-01", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Oyegoke Oyebode", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I4210148469", "display_name": "Visa (United States)", "ror": "https://ror.org/05t1y0b59", "country_code": "US", "type": "company", "lineage": ["https://openalex.org/I4210148469"]}], "countries": ["US"], "is_corresponding": true, "raw_author_name": "Oyegoke Oyebode", "raw_affiliation_strings": ["Technical Program Manager , Visa Inc . USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Technical Program Manager , Visa Inc . USA", "institution_ids": ["https://openalex.org/I4210148469"]}]}], "abstract_inverted_index": {"The": [0], "deployment": [1], "of": [2, 211], "artificial": [3], "intelligence": [4], "in": [5, 18, 61, 100, 170, 189], "healthcare": [6, 93, 140], "is": [7], "increasingly": [8], "constrained": [9], "by": [10, 29, 46, 82, 163], "privacy,": [11, 206], "equity,": [12], "and": [13, 20, 38, 59, 91, 104, 118, 185, 196, 207, 215, 223], "regulatory": [14], "compliance": [15], "challenges,": [16], "especially": [17], "multilingual": [19, 92, 165, 193], "cross-border": [21], "contexts.Traditional": [22], "centralized": [23], "machine": [24], "learning": [25, 41], "approaches": [26], "are": [27, 107], "limited": [28], "restrictions": [30], "on": [31, 111], "patient": [32, 136, 232], "data": [33], "sharing,": [34], "raising": [35], "both": [36], "ethical": [37], "legal": [39], "concerns.Federated": [40], "offers": [42], "a": [43, 65, 70, 122, 219], "promising": [44], "solution": [45], "enabling": [47, 164, 227], "distributed": [48], "training": [49], "across": [50, 182], "institutions": [51], "without": [52, 133], "transferring": [53], "raw": [54], "data,": [55], "yet": [56], "ensuring": [57], "trust": [58, 124], "privacy": [60, 156], "federated": [62, 79, 113, 190, 213], "systems": [63], "remains": [64], "critical": [66], "barrier.This": [67], "study": [68], "proposes": [69], "novel": [71], "framework": [72, 158], "that": [73, 147, 167], "combines": [74], "transformer": [75], "architectures": [76], "with": [77], "encrypted": [78, 112], "datasets": [80, 114], "anchored": [81], "blockchain": [83, 216], "zero-knowledge": [84, 127], "proofs": [85, 128], "(ZKPs)": [86], "to": [87, 109, 142, 154, 203], "achieve": [88], "privacy-preserving,": [89], "equitable,": [90], "diagnostics.Transformer-based": [94], "models,": [95], "known": [96], "for": [97], "their": [98], "strength": [99], "natural": [101], "language": [102], "processing": [103], "multimodal": [105], "learning,": [106, 214], "adapted": [108], "operate": [110], "spanning": [115], "diverse": [116, 183], "linguistic": [117, 186], "demographic": [119], "contexts.Blockchain": [120], "provides": [121], "decentralized": [123], "layer,": [125], "while": [126, 152, 230], "ensure": [129], "verifiable": [130], "model": [131, 180], "updates": [132], "exposing": [134], "sensitive": [135], "information.This": [137], "combination": [138], "allows": [139], "providers": [141], "collaboratively": [143], "train": [144], "diagnostic": [145], "models": [146], "maintain": [148], "strong": [149], "predictive": [150], "performance": [151], "adhering": [153], "strict": [155], "guarantees.The": [157], "also": [159], "advances": [160], "health": [161], "equity": [162], "diagnostics": [166, 198], "address": [168], "disparities": [169], "underrepresented": [171], "populations.By": [172], "integrating": [173], "explainability": [174], "mechanisms,": [175], "stakeholders": [176], "gain": [177], "insights": [178], "into": [179], "reasoning": [181], "cultural": [184], "datasets.Case": [187], "applications": [188], "medical": [191], "imaging,": [192], "clinical": [194], "notes,": [195], "genomic": [197], "highlight": [199], "the": [200, 209], "framework's": [201], "capacity": [202], "balance": [204], "accuracy,": [205], "fairness.Overall,": [208], "integration": [210], "transformers,": [212], "ZKPs": [217], "represents": [218], "pathway": [220], "toward": [221], "trustworthy": [222], "equitable": [224], "AI-driven": [225], "healthcare,": [226], "collaborative": [228], "innovation": [229], "safeguarding": [231], "rights.": [233]}, "primary_topic": {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.9868000149726868, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.9868000149726868, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10270", "display_name": "Blockchain Technology Applications and Security", "score": 0.011300000362098217, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11636", "display_name": "Artificial Intelligence in Healthcare and Education", "score": 0.0010000000474974513, "subfield": {"id": "https://openalex.org/subfields/2718", "display_name": "Health Informatics"}, "field": {"id": "https://openalex.org/fields/27", "display_name": "Medicine"}, "domain": {"id": "https://openalex.org/domains/4", "display_name": "Health Sciences"}}], "primary_location": {"id": "doi:10.55248/gengpi.6.0825.3115", "is_oa": true, "landing_page_url": "https://doi.org/10.55248/gengpi.6.0825.3115", "pdf_url": "https://doi.org/10.55248/gengpi.6.0825.3115", "source": {"id": "https://openalex.org/S4220651630", "display_name": "International Journal of Research Publication and Reviews", "issn_l": "2582-7421", "issn": ["2582-7421"], "is_oa": false, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": null, "host_organization_name": null, "host_organization_lineage": [], "host_organization_lineage_names": [], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "International Journal of Research Publication and Reviews", "raw_type": "journal-article"}, "cited_by_count": 1, "is_retracted": false}, {"id": "https://openalex.org/W7117146154", "doi": "https://doi.org/10.48550/arxiv.2512.18127", "title": "ACE-Sync: An Adaptive Cloud-Edge Synchronization Framework for Communication-Efficient Large-Scale Distributed Model Training", "publication_year": 2025, "publication_date": "2025-12-19", "type": "preprint", "language": null, "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5010277474", "display_name": "Yi Yang", "orcid": "https://orcid.org/0000-0001-7797-6341", "observed_orcids": ["https://orcid.org/0000-0001-7797-6341"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Yang, Yi", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5121244465", "display_name": "Ziyu Lin", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Lin, Ziyu", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5121142467", "display_name": "Liesheng Wei", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Wei, Liesheng", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Large-scale": [0], "deep": [1], "learning": [2], "models": [3], "impose": [4], "substantial": [5], "communication": [6, 30, 129, 141, 163], "overh": [7], "ead": [8], "in": [9, 13], "distributed": [10, 186, 204], "training,": [11], "particularly": [12], "bandwidth-constrained": [14], "or": [15, 22], "heterogeneous": [16], "clo": [17], "ud-edge": [18], "environments.": [19], "Conventional": [20], "synchronous": [21], "fixed-compression": [23], "techniques": [24], "o": [25], "ften": [26], "struggle": [27], "to": [28, 80, 98, 146, 157], "balance": [29], "cost,": [31], "convergence": [32, 154], "stability,": [33], "and": [34, 66, 82, 112, 119, 152, 182, 197], "model": [35, 168, 205], "accura": [36], "cy.": [37], "To": [38], "address": [39], "these": [40], "challenges,": [41], "we": [42], "propose": [43], "ACE-Sync,": [44], "an": [45, 54], "Adaptive": [46], "Cloud-Edge": [47], "Sy": [48], "nchronization": [49], "Framework": [50], "that": [51, 124, 191], "integrates": [52], "(1)": [53], "attention-based": [55], "gradient": [56, 101], "importance": [57], "p": [58], "redictor,": [59], "(2)": [60], "a": [61, 68, 194], "differentiated": [62], "parameter": [63, 78], "compression": [64, 85], "strategy,": [65], "(3)": [67], "hierarchical": [69], "cl": [70], "oud-edge": [71], "coordination": [72], "mechanism.": [73], "ACE-Sync": [74, 125, 139, 165, 192], "dynamically": [75], "selects": [76], "which": [77], "groups": [79], "synchronize": [81], "determines": [83], "appropriate": [84], "levels": [86], "under": [87], "per-devic": [88], "e": [89, 162], "bandwidth": [90], "budgets.": [91], "A": [92], "knapsack-based": [93], "optimization": [94], "strategy": [95], "is": [96], "adopted": [97], "maximize": [99], "important": [100], "preservation": [102], "while": [103, 131], "reducing": [104], "redundant": [105], "communication.": [106], "Furthermo": [107], "re,": [108], "residual-based": [109], "error": [110], "compensation": [111], "device": [113], "clustering": [114], "ensure": [115], "long-term": [116], "converg": [117], "ence": [118], "cross-device": [120], "personalization.": [121], "Experiments": [122], "show": [123], "substantiall": [126], "y": [127], "reduces": [128], "overhead": [130], "maintaining": [132], "competitive": [133], "accuracy.": [134], "Compar": [135], "ed": [136], "with": [137], "FullSync,": [138], "lowers": [140], "cost": [142], "from": [143, 155], "112.5": [144], "GB": [145, 148], "44.7": [147], "(a": [149], "60%": [150], "reduction)": [151], "shortens": [153], "41": [156], "39": [158], "epochs.": [159], "Despite": [160], "aggressiv": [161], "reduction,": [164], "preserves": [166], "high": [167], "quality,": [169], "achieving": [170], "82.": [171], "1%": [172], "Top-1": [173], "accuracy-only": [174], "0.3%": [175], "below": [176], "the": [177], "full-synchronization": [178], "baseline-demonstrating": [179], "its": [180], "efficiency": [181], "scalability": [183], "for": [184, 201], "large-scale": [185, 202], "training.": [187, 206], "These": [188], "results": [189], "indicate": [190], "provides": [193], "scalable,": [195], "communication-efficient,": [196], "accuracy-preservin": [198], "g": [199], "solution": [200], "cloud-edge": [203]}, "primary_topic": {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.8690999746322632, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.8690999746322632, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.018400000408291817, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.014499999582767487, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2512.18127", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2512.18127", "pdf_url": "https://arxiv.org/pdf/2512.18127", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4410056820", "doi": "https://doi.org/10.55041/ijsrem46707", "title": "Scalable Streaming Data at the Egde Using Deep Learning", "publication_year": 2025, "publication_date": "2025-05-03", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5104267550", "display_name": "P. Kamakshi Thai", "orcid": "https://orcid.org/0009-0004-8361-4741", "observed_orcids": ["https://orcid.org/0009-0004-8361-4741"]}, "institutions": [], "countries": [], "is_corresponding": true, "raw_author_name": "P. Kamakshi Thai", "raw_affiliation_strings": ["Assistant Professor of Department of CSE (AI & ML) of ACE Engineering College , India"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Assistant Professor of Department of CSE (AI & ML) of ACE Engineering College , India", "institution_ids": []}]}], "abstract_inverted_index": {"ABSTRACT": [0], "In": [1], "recent": [2], "years,": [3], "the": [4, 122, 129, 134], "proliferation": [5], "of": [6, 124], "IoT": [7], "devices": [8], "and": [9, 63, 65, 90, 107, 142], "edge": [10, 47, 97, 130], "sensors": [11], "has": [12], "led": [13], "to": [14, 53, 93, 112], "an": [15], "exponential": [16], "increase": [17], "in": [18, 70, 138], "real-time": [19, 125], "streaming": [20, 59, 88], "data,": [21, 89], "challenging": [22], "conventional": [23], "deep": [24, 42, 115], "learning": [25, 43, 116], "methods": [26], "designed": [27], "for": [28, 40, 81, 86, 136], "static,": [29], "centralized": [30], "datasets.": [31], "ScaDLES,": [32], "addresses": [33], "these": [34], "challenges": [35], "by": [36], "introducing": [37], "a": [38, 95], "framework": [39], "training": [41, 105], "models": [44], "directly": [45], "on": [46, 128], "devices.": [48], "The": [49, 73], "primary": [50], "objective": [51], "is": [52, 75], "overcome": [54], "issues": [55], "such": [56], "as": [57], "heterogeneous": [58], "rates,": [60], "limited": [61], "memory": [62], "storage,": [64], "high": [66], "communication": [67], "overhead": [68], "typical": [69], "distributed": [71, 96, 114], "environments.": [72], "system": [74], "implemented": [76], "using": [77], "Python": [78], "with": [79], "PyTorch": [80], "model": [82, 126], "training,": [83], "Apache": [84], "Kafka": [85], "simulating": [87], "Docker": [91], "containers": [92], "emulate": [94], "environment.": [98], "it": [99], "indicates": [100], "that": [101], "ScaDLES": [102], "significantly": [103], "enhances": [104], "speed": [106], "reduces": [108], "resource": [109], "consumption": [110], "compared": [111], "traditional": [113], "frameworks.": [117], "It\u2019s": [118], "not": [119], "only": [120], "improves": [121], "feasibility": [123], "updates": [127], "but": [131], "also": [132], "paves": [133], "way": [135], "applications": [137], "surveillance,": [139], "autonomous": [140], "vehicles,": [141], "smart": [143], "cities.": [144], "Keywords:": [145], "Edge": [146], "Computing,": [147], "Streaming": [148], "Data,": [149], "Deep": [150], "Learning,": [151], "Distributed": [152], "Training": [153], "Gradient-": [154], "Compression.": [155]}, "primary_topic": {"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.9028000235557556, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.9028000235557556, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T12761", "display_name": "Data Stream Mining Techniques", "score": 0.08399999886751175, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10317", "display_name": "Advanced Database Systems and Queries", "score": 0.0015999999595806003, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.55041/ijsrem46707", "is_oa": true, "landing_page_url": "https://doi.org/10.55041/ijsrem46707", "pdf_url": "https://ijsrem.com/download/scalable-streaming-data-at-the-egde-using-deep-learning/?wpdmdl=50949&refresh=681712b9541fa1746342585", "source": {"id": "https://openalex.org/S4210240841", "display_name": "INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT", "issn_l": "2582-3930", "issn": ["2582-3930"], "is_oa": false, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": null, "host_organization_name": null, "host_organization_lineage": [], "host_organization_lineage_names": [], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "INTERNATIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT", "raw_type": "journal-article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4417136570", "doi": "https://doi.org/10.48550/arxiv.2506.07998", "title": "Generative Modeling of Weights: Generalization or Memorization?", "publication_year": 2025, "publication_date": "2025-06-09", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5112871434", "display_name": "Boya Zeng", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Zeng, Boya", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": null, "display_name": "Yin, Yida", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Yin, Yida", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": null, "display_name": "Xu, Zhiqiu", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Xu, Zhiqiu", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5100452089", "display_name": "Zhuang Liu", "orcid": "https://orcid.org/0000-0001-6395-0212", "observed_orcids": ["https://orcid.org/0000-0001-6395-0212", "https://orcid.org/0000-0002-4695-6345"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Liu, Zhuang", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Generative": [0], "models": [1, 150], "have": [2], "recently": [3], "been": [4], "explored": [5], "for": [6, 141], "synthesizing": [7], "neural": [8, 14], "network": [9, 15], "weights.": [10], "These": [11, 136], "approaches": [12], "take": [13], "checkpoints": [16, 52], "as": [17, 93], "training": [18, 83], "data": [19], "and": [20, 108, 126, 145], "aim": [21], "to": [22, 40, 57, 88, 96, 133, 153], "generate": [23, 41], "high-performing": [24, 110], "weights": [25, 46, 68, 98], "during": [26, 54], "inference.": [27], "In": [28], "this": [29, 117], "work,": [30, 61], "we": [31, 62], "examine": [32], "four": [33], "representative,": [34], "well-known": [35], "methods": [36, 66], "on": [37], "their": [38], "ability": [39], "novel": [42], "model": [43], "weights,": [44], "i.e.,": [45], "that": [47, 64, 116], "are": [48], "different": [49, 107], "from": [50, 121], "the": [51, 82, 97, 127, 139], "seen": [53], "training.": [55], "Contrary": [56], "claims": [58], "in": [59, 105], "prior": [60], "find": [63], "these": [65], "synthesize": [67], "largely": [69], "by": [70], "memorization:": [71], "they": [72, 86], "produce": [73], "either": [74], "replicas,": [75], "or,": [76], "at": [77, 160], "best,": [78], "simple": [79, 90, 102], "interpolations": [80], "of": [81, 129, 148], "checkpoints.": [84], "Moreover,": [85], "fail": [87], "outperform": [89], "baselines,": [91], "such": [92], "adding": [94], "noise": [95], "or": [99], "taking": [100], "a": [101], "weight": [103, 134], "ensemble,": [104], "obtaining": [106], "simultaneously": [109], "models.": [111], "Our": [112, 156], "further": [113], "analysis": [114], "suggests": [115], "memorization": [118], "might": [119], "result": [120], "limited": [122], "data,": [123], "overparameterized": [124], "models,": [125], "underuse": [128], "structural": [130], "priors": [131], "specific": [132], "data.": [135], "findings": [137], "highlight": [138], "need": [140], "more": [142], "careful": [143], "design": [144], "rigorous": [146], "evaluation": [147], "generative": [149], "when": [151], "applied": [152], "new": [154], "domains.": [155], "code": [157], "is": [158], "available": [159], "https://github.com/boyazeng/weight_memorization.": [161]}, "primary_topic": {"id": "https://openalex.org/T10775", "display_name": "Generative Adversarial Networks and Image Synthesis", "score": 0.35600000619888306, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10775", "display_name": "Generative Adversarial Networks and Image Synthesis", "score": 0.35600000619888306, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11307", "display_name": "Domain Adaptation and Few-Shot Learning", "score": 0.1597999930381775, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.1527000069618225, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2506.07998", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2506.07998", "pdf_url": "https://arxiv.org/pdf/2506.07998", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4417526447", "doi": "https://doi.org/10.62306/7ds08r28", "title": "A Survey on Integrated Training-Inference Architectures for Large Language Models on Multi-GPU Stream Processors", "publication_year": 2025, "publication_date": "2025-12-20", "type": "article", "language": null, "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Heric Tsang", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": true, "raw_author_name": "Heric Tsang", "raw_affiliation_strings": ["( . Sourcera Science and Technology Co. , Limited Hong Kong 999077"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "( . Sourcera Science and Technology Co. , Limited Hong Kong 999077", "institution_ids": []}]}], "abstract_inverted_index": {"Large": [0], "language": [1, 13], "models": [2], "(LLMs)": [3], "have": [4], "revolutionized": [5], "artificial": [6], "intelligence,": [7], "achieving": [8], "remarkable": [9], "performance": [10], "in": [11, 33, 112, 123, 165], "natural": [12], "understanding,": [14], "generation,": [15], "and": [16, 35, 71, 88, 101, 109, 126, 140, 168], "multimodal": [17], "tasks.": [18], "However,": [19], "their": [20], "unprecedented": [21], "scale\u2014often": [22], "comprising": [23], "billions": [24], "to": [25, 105], "trillions": [26], "of": [27, 52, 80], "parameters\u2014imposes": [28], "severe": [29], "computational": [30], "demands,": [31], "particularly": [32], "training": [34], "inference": [36], "phases,": [37], "necessitating": [38], "advanced": [39], "parallel": [40], "processing": [41], "architectures": [42, 56], "on": [43, 59], "multi-GPU": [44, 65], "arrays.": [45], "This": [46], "survey": [47, 150], "provides": [48], "a": [49], "comprehensive": [50], "overview": [51], "integrated": [53], "training-inference": [54], "(train-infer)": [55], "for": [57, 129, 154, 162], "LLMs": [58], "large-scale": [60], "GPU": [61], "stream": [62, 66], "processors,": [63], "emphasizing": [64], "processing,": [67], "hypercube": [68], "tensor": [69, 89, 103], "parallelism,": [70, 85, 87, 90], "hardware-software": [72], "co-designed": [73], "compilation": [74, 128], "frameworks.": [75], "We": [76], "trace": [77], "the": [78, 160], "evolution": [79], "parallelism": [81], "strategies,": [82], "including": [83, 121], "data": [84], "pipeline": [86, 96], "highlighting": [91], "innovations": [92], "such": [93], "as": [94], "cross-cluster": [95], "execution,": [97], "adaptive": [98], "NIC": [99], "selection,": [100], "spatiotemporal": [102], "partitioning": [104], "mitigate": [106], "communication": [107], "overheads": [108], "memory": [110], "bottlenecks": [111], "heterogeneous": [113], "environments": [114], "[1,": [115], "7,": [116], "8,": [117], "17].": [118], "Key": [119], "challenges,": [120], "scalability": [122], "non-homogeneous": [124], "networks": [125], "efficient": [127], "diverse": [130], "hardware,": [131], "are": [132], "analyzed": [133], "alongside": [134], "state-of-the-art": [135], "solutions": [136], "like": [137], "MLIR-based": [138], "frameworks": [139], "RISC-V": [141], "accelerators": [142], "[28,": [143], "33].": [144], "By": [145], "synthesizing": [146], "recent": [147], "advancements,": [148], "this": [149], "identifies": [151], "promising": [152], "directions": [153], "scalable,": [155], "energy-efficient": [156], "LLM": [157], "systems,": [158], "paving": [159], "way": [161], "broader": [163], "deployment": [164], "edge": [166], "computing": [167], "high-performance": [169], "clusters.": [170]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9955999851226807, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9955999851226807, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.0006000000284984708, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.0005000000237487257, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.62306/7ds08r28", "is_oa": true, "landing_page_url": "https://doi.org/10.62306/7ds08r28", "pdf_url": "https://d-sci.org/index.php/dsci/article/download/18/16", "source": {"id": "https://openalex.org/S5407052603", "display_name": "Digital Science", "issn_l": "3006-0753", "issn": ["3006-0753", "3006-2942"], "is_oa": false, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": null, "host_organization_name": null, "host_organization_lineage": [], "host_organization_lineage_names": [], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Digital Science", "raw_type": "journal-article"}, "cited_by_count": 1, "is_retracted": false}, {"id": "https://openalex.org/W4417150036", "doi": "https://doi.org/10.48550/arxiv.2503.16585", "title": "Distributed LLMs and Multimodal Large Language Models: A Survey on Advances, Challenges, and Future Directions", "publication_year": 2025, "publication_date": "2025-03-20", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Amini, Hadi", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Amini, Hadi", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5090885879", "display_name": "Md. Jueal Mia", "orcid": "https://orcid.org/0000-0002-9383-8739", "observed_orcids": ["https://orcid.org/0000-0002-9383-8739"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Mia, Md Jueal", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5004311783", "display_name": "Yasaman Saadati", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Saadati, Yasaman", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5077711449", "display_name": "Ahmed Imteaj", "orcid": "https://orcid.org/0000-0002-6975-3997", "observed_orcids": ["https://orcid.org/0000-0002-6975-3997"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Imteaj, Ahmed", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5043768774", "display_name": "Seyedsina Nabavirazavi", "orcid": "https://orcid.org/0000-0002-9186-4386", "observed_orcids": ["https://orcid.org/0000-0002-9186-4386"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Nabavirazavi, Seyedsina", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5090396089", "display_name": "Urmish Thakker", "orcid": "https://orcid.org/0000-0002-0515-9155", "observed_orcids": ["https://orcid.org/0000-0002-0515-9155"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Thakker, Urmish", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5071571852", "display_name": "Md Zarif Hossain", "orcid": "https://orcid.org/0009-0007-0180-6371", "observed_orcids": ["https://orcid.org/0009-0007-0180-6371"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Hossain, Md Zarif", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5080595441", "display_name": "Awal Ahmed Fime", "orcid": "https://orcid.org/0009-0004-0015-9297", "observed_orcids": ["https://orcid.org/0009-0004-0015-9297"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Fime, Awal Ahmed", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5102765260", "display_name": "Sitharama S. Iyengar", "orcid": "https://orcid.org/0000-0003-3203-833X", "observed_orcids": ["https://orcid.org/0000-0003-3203-833X"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Iyengar, S. S.", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Language": [0], "models": [1, 6, 129, 133, 141], "(LMs)": [2], "are": [3, 152], "machine": [4, 42], "learning": [5], "designed": [7, 153], "to": [8, 56, 100, 154, 166, 239], "predict": [9], "linguistic": [10], "patterns": [11], "by": [12], "estimating": [13], "the": [14, 74, 79, 181, 194, 205, 234, 241], "probability": [15], "of": [16, 31, 81, 158, 200, 213, 245], "word": [17], "sequences": [18], "based": [19, 207], "on": [20, 96, 120, 146, 208], "large-scale": [21], "datasets,": [22], "such": [23], "as": [24], "text.": [25], "LMs": [26, 227], "have": [27], "a": [28, 53, 118], "wide": [29], "range": [30], "applications": [32], "in": [33, 58, 84, 219], "natural": [34], "language": [35, 128, 132, 140], "processing": [36, 147], "(NLP)": [37], "tasks,": [38], "including": [39, 126, 184], "autocomplete": [40], "and": [41, 61, 72, 86, 104, 111, 138, 148, 163, 165, 189, 197, 228, 243], "translation.": [43], "Although": [44], "larger": [45], "datasets": [46, 83], "typically": [47], "enhance": [48, 240], "LM": [49], "performance,": [50], "scalability": [51, 71], "remains": [52], "challenge": [54], "due": [55], "constraints": [57], "computational": [59, 76, 109], "power": [60], "resources.": [62], "Distributed": [63], "computing": [64], "strategies": [65], "offer": [66], "essential": [67], "solutions": [68, 122, 225, 238], "for": [69, 123, 169, 222, 226, 236], "improving": [70], "managing": [73], "growing": [75], "demand.": [77], "Further,": [78, 202], "use": [80], "sensitive": [82], "training": [85, 103], "deployment": [87], "raises": [88], "significant": [89], "privacy": [90], "concerns.": [91], "Recent": [92], "research": [93, 231], "has": [94], "focused": [95], "developing": [97], "decentralized": [98], "techniques": [99], "enable": [101], "distributed": [102, 121, 185, 224, 246], "inference": [105], "while": [106, 191], "utilizing": [107], "diverse": [108], "resources": [110], "enabling": [112, 223], "edge": [113], "AI.": [114], "This": [115], "paper": [116, 176], "presents": [117], "survey": [119], "various": [124], "LMs,": [125], "large": [127], "(LLMs),": [130], "vision": [131], "(VLMs),": [134], "multimodal": [135], "LLMs": [136, 144], "(MLLMs),": [137], "small": [139], "(SLMs).": [142], "While": [143], "focus": [145, 211], "generating": [149], "text,": [150, 161], "MLLMs": [151], "handle": [155], "multiple": [156], "modalities": [157], "data": [159], "(e.g.,": [160], "images,": [162], "audio)": [164], "integrate": [167], "them": [168], "broader": [170], "applications.": [171], "To": [172], "this": [173, 175], "end,": [174], "reviews": [177], "key": [178], "advancements": [179], "across": [180], "MLLM": [182], "pipeline,": [183], "training,": [186], "inference,": [187], "fine-tuning,": [188], "deployment,": [190], "also": [192], "identifying": [193], "contributions,": [195], "limitations,": [196], "future": [198, 230], "areas": [199, 212], "improvement.": [201], "it": [203], "categorizes": [204], "literature": [206], "six": [209], "primary": [210], "decentralization.": [214], "Our": [215], "analysis": [216], "describes": [217], "gaps": [218], "current": [220], "methodologies": [221], "outline": [229], "directions,": [232], "emphasizing": [233], "need": [235], "novel": [237], "robustness": [242], "applicability": [244], "LMs.": [247]}, "primary_topic": {"id": "https://openalex.org/T11714", "display_name": "Multimodal Machine Learning Applications", "score": 0.5938000082969666, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T11714", "display_name": "Multimodal Machine Learning Applications", "score": 0.5938000082969666, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.19059999287128448, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10181", "display_name": "Natural Language Processing Techniques", "score": 0.08510000258684158, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2503.16585", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2503.16585", "pdf_url": "https://arxiv.org/pdf/2503.16585", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4409581188", "doi": "https://doi.org/10.1109/jiot.2025.3558933", "title": "Resource Allocation in Flexible-Bandwidth Fine-Grained Optical Transport Networks for Geo-Distributed Machine Learning", "publication_year": 2025, "publication_date": "2025-04-18", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5100527016", "display_name": "Meng Lian", "orcid": "https://orcid.org/0000-0002-2202-0454", "observed_orcids": ["https://orcid.org/0000-0002-2202-0454"]}, "institutions": [{"id": "https://openalex.org/I139759216", "display_name": "Beijing University of Posts and Telecommunications", "ror": "https://ror.org/04w9fbh59", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I139759216"]}, {"id": "https://openalex.org/I4391767763", "display_name": "State Key Laboratory of Information Photonics and Optical Communications", "ror": "https://ror.org/02v9wg343", "country_code": null, "type": "facility", "lineage": ["https://openalex.org/I139759216", "https://openalex.org/I4391767763"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Meng Lian", "raw_affiliation_strings": ["State Key Laboratory of Information Photonic and Optical Communication, Beijing University of Posts and Telecommunications, Beijing, China", "Beijing University of Posts and Telecommunications, Beijing, China"], "raw_orcid": "https://orcid.org/0000-0002-2202-0454", "affiliations": [{"raw_affiliation_string": "State Key Laboratory of Information Photonic and Optical Communication, Beijing University of Posts and Telecommunications, Beijing, China", "institution_ids": ["https://openalex.org/I139759216", "https://openalex.org/I4391767763"]}, {"raw_affiliation_string": "Beijing University of Posts and Telecommunications, Beijing, China", "institution_ids": ["https://openalex.org/I139759216"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5030811766", "display_name": "Yongli Zhao", "orcid": "https://orcid.org/0000-0003-3716-8248", "observed_orcids": ["https://orcid.org/0000-0003-3716-8248"]}, "institutions": [{"id": "https://openalex.org/I139759216", "display_name": "Beijing University of Posts and Telecommunications", "ror": "https://ror.org/04w9fbh59", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I139759216"]}, {"id": "https://openalex.org/I4391767763", "display_name": "State Key Laboratory of Information Photonics and Optical Communications", "ror": "https://ror.org/02v9wg343", "country_code": null, "type": "facility", "lineage": ["https://openalex.org/I139759216", "https://openalex.org/I4391767763"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yongli Zhao", "raw_affiliation_strings": ["State Key Laboratory of Information Photonic and Optical Communication, Beijing University of Posts and Telecommunications, Beijing, China", "Beijing University of Posts and Telecommunications, Beijing, China"], "raw_orcid": "https://orcid.org/0000-0003-3716-8248", "affiliations": [{"raw_affiliation_string": "State Key Laboratory of Information Photonic and Optical Communication, Beijing University of Posts and Telecommunications, Beijing, China", "institution_ids": ["https://openalex.org/I139759216", "https://openalex.org/I4391767763"]}, {"raw_affiliation_string": "Beijing University of Posts and Telecommunications, Beijing, China", "institution_ids": ["https://openalex.org/I139759216"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5101549985", "display_name": "Xin Li", "orcid": "https://orcid.org/0000-0002-9388-4241", "observed_orcids": ["https://orcid.org/0000-0002-9388-4241"]}, "institutions": [{"id": "https://openalex.org/I139759216", "display_name": "Beijing University of Posts and Telecommunications", "ror": "https://ror.org/04w9fbh59", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I139759216"]}, {"id": "https://openalex.org/I4391767763", "display_name": "State Key Laboratory of Information Photonics and Optical Communications", "ror": "https://ror.org/02v9wg343", "country_code": null, "type": "facility", "lineage": ["https://openalex.org/I139759216", "https://openalex.org/I4391767763"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Xin Li", "raw_affiliation_strings": ["State Key Laboratory of Information Photonic and Optical Communication, Beijing University of Posts and Telecommunications, Beijing, China", "Beijing University of Posts and Telecommunications, Beijing, China"], "raw_orcid": "https://orcid.org/0000-0002-9388-4241", "affiliations": [{"raw_affiliation_string": "State Key Laboratory of Information Photonic and Optical Communication, Beijing University of Posts and Telecommunications, Beijing, China", "institution_ids": ["https://openalex.org/I139759216", "https://openalex.org/I4391767763"]}, {"raw_affiliation_string": "Beijing University of Posts and Telecommunications, Beijing, China", "institution_ids": ["https://openalex.org/I139759216"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Wenhong Liu", "orcid": "https://orcid.org/0009-0006-2892-6704", "observed_orcids": ["https://orcid.org/0009-0006-2892-6704"]}, "institutions": [{"id": "https://openalex.org/I139759216", "display_name": "Beijing University of Posts and Telecommunications", "ror": "https://ror.org/04w9fbh59", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I139759216"]}, {"id": "https://openalex.org/I4391767763", "display_name": "State Key Laboratory of Information Photonics and Optical Communications", "ror": "https://ror.org/02v9wg343", "country_code": null, "type": "facility", "lineage": ["https://openalex.org/I139759216", "https://openalex.org/I4391767763"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Wenhong Liu", "raw_affiliation_strings": ["State Key Laboratory of Information Photonic and Optical Communication, Beijing University of Posts and Telecommunications, Beijing, China", "Beijing University of Posts and Telecommunications, Beijing, China"], "raw_orcid": "https://orcid.org/0009-0006-2892-6704", "affiliations": [{"raw_affiliation_string": "State Key Laboratory of Information Photonic and Optical Communication, Beijing University of Posts and Telecommunications, Beijing, China", "institution_ids": ["https://openalex.org/I139759216", "https://openalex.org/I4391767763"]}, {"raw_affiliation_string": "Beijing University of Posts and Telecommunications, Beijing, China", "institution_ids": ["https://openalex.org/I139759216"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100700731", "display_name": "Yajie Li", "orcid": "https://orcid.org/0000-0002-8751-7947", "observed_orcids": ["https://orcid.org/0000-0002-8751-7947"]}, "institutions": [{"id": "https://openalex.org/I139759216", "display_name": "Beijing University of Posts and Telecommunications", "ror": "https://ror.org/04w9fbh59", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I139759216"]}, {"id": "https://openalex.org/I4391767763", "display_name": "State Key Laboratory of Information Photonics and Optical Communications", "ror": "https://ror.org/02v9wg343", "country_code": null, "type": "facility", "lineage": ["https://openalex.org/I139759216", "https://openalex.org/I4391767763"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yajie Li", "raw_affiliation_strings": ["State Key Laboratory of Information Photonic and Optical Communication, Beijing University of Posts and Telecommunications, Beijing, China", "Beijing University of Posts and Telecommunications, Beijing, China"], "raw_orcid": "https://orcid.org/0000-0002-8751-7947", "affiliations": [{"raw_affiliation_string": "State Key Laboratory of Information Photonic and Optical Communication, Beijing University of Posts and Telecommunications, Beijing, China", "institution_ids": ["https://openalex.org/I139759216", "https://openalex.org/I4391767763"]}, {"raw_affiliation_string": "Beijing University of Posts and Telecommunications, Beijing, China", "institution_ids": ["https://openalex.org/I139759216"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5038468421", "display_name": "Massimo Tornatore", "orcid": "https://orcid.org/0000-0003-0740-1061", "observed_orcids": ["https://orcid.org/0000-0003-0740-1061"]}, "institutions": [{"id": "https://openalex.org/I93860229", "display_name": "Politecnico di Milano", "ror": "https://ror.org/01nffqt88", "country_code": "IT", "type": "education", "lineage": ["https://openalex.org/I93860229"]}], "countries": ["IT"], "is_corresponding": false, "raw_author_name": "Massimo Tornatore", "raw_affiliation_strings": ["Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Milan, Italy"], "raw_orcid": "https://orcid.org/0000-0003-0740-1061", "affiliations": [{"raw_affiliation_string": "Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Milan, Italy", "institution_ids": ["https://openalex.org/I93860229"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5100436639", "display_name": "Jie Zhang", "orcid": "https://orcid.org/0000-0001-7750-2197", "observed_orcids": ["https://orcid.org/0000-0001-7750-2197", "https://orcid.org/0000-0001-9638-574X", "https://orcid.org/0009-0008-4670-9038"]}, "institutions": [{"id": "https://openalex.org/I139759216", "display_name": "Beijing University of Posts and Telecommunications", "ror": "https://ror.org/04w9fbh59", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I139759216"]}, {"id": "https://openalex.org/I4391767763", "display_name": "State Key Laboratory of Information Photonics and Optical Communications", "ror": "https://ror.org/02v9wg343", "country_code": null, "type": "facility", "lineage": ["https://openalex.org/I139759216", "https://openalex.org/I4391767763"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Jie Zhang", "raw_affiliation_strings": ["State Key Laboratory of Information Photonic and Optical Communication, Beijing University of Posts and Telecommunications, Beijing, China", "Beijing University of Posts and Telecommunications, Beijing, China"], "raw_orcid": "https://orcid.org/0000-0001-7750-2197", "affiliations": [{"raw_affiliation_string": "State Key Laboratory of Information Photonic and Optical Communication, Beijing University of Posts and Telecommunications, Beijing, China", "institution_ids": ["https://openalex.org/I139759216", "https://openalex.org/I4391767763"]}, {"raw_affiliation_string": "Beijing University of Posts and Telecommunications, Beijing, China", "institution_ids": ["https://openalex.org/I139759216"]}]}], "abstract_inverted_index": {"Geo-distributed": [0], "machine": [1], "learning": [2, 7, 110], "(GDML)": [3], "can": [4], "facilitate": [5], "collaborative": [6], "among": [8, 85, 104], "geographically-dispersed": [9], "data": [10], "centers": [11], "to": [12, 57, 90, 108, 192], "meet": [13], "the": [14, 29, 68, 135, 197], "demands": [15], "of": [16, 25, 31, 138], "distributed": [17, 23, 32, 42], "and": [18, 66, 82, 163, 173], "privacy-preserving": [19], "training": [20, 33], "for": [21, 70, 93, 119, 154, 189], "large-scale": [22], "Internet": [24], "Things": [26], "applications.": [27], "Unfortunately,": [28], "efficiency": [30], "tasks": [34, 75, 86, 191], "heavily": [35], "depends": [36], "on": [37, 150, 170, 177], "synchronized": [38], "communication": [39], "between": [40], "multiple": [41], "models": [43], "over": [44], "bandwidth-limited": [45], "wide": [46], "area": [47], "networks": [48], "(WANs).": [49], "The": [50], "fine-grained": [51], "Optical": [52], "Transport": [53], "Network": [54], "(fgOTN),": [55], "thanks": [56], "its": [58], "adjustable": [59], "bandwidth": [60, 80, 99, 164], "connections,": [61], "represents": [62], "more": [63], "flexible": [64, 79, 98], "transmission": [65], "has": [67], "ability": [69], "accurate": [71], "synchronization": [72], "across": [73], "GDML": [74, 94, 120, 139, 155], "in": [76, 95, 121, 156], "WANs.": [77], "However,": [78], "assignment": [81, 100], "complex": [83], "interdependencies": [84], "pose": [87], "significant": [88], "challenges": [89], "resource": [91, 102, 116, 146, 187, 194], "allocation": [92, 117, 147, 188], "fgOTN.": [96, 122, 157], "Specifically,": [97], "exacerbates": [101], "competition": [103], "task": [105, 161, 199], "flows,": [106], "leading": [107], "decreased": [109], "efficiency.": [111], "This": [112], "paper": [113], "provides": [114], "novel": [115], "solutions": [118], "We": [123], "first": [124], "formulate": [125], "this": [126], "problem": [127], "as": [128], "a": [129], "linear": [130], "programming": [131], "aimed": [132], "at": [133], "maximizing": [134], "completion": [136, 162, 178, 200], "ratio": [137, 201], "tasks.": [140], "Subsequently,": [141], "we": [142], "propose": [143], "an": [144], "innovative": [145], "algorithm": [148, 152], "based": [149, 169, 176], "genetic": [151], "(GARA)": [153], "GARA": [158, 184], "considers": [159], "both": [160], "adjustment": [165], "through": [166], "population": [167], "generation": [168], "prior": [171], "knowledge": [172], "adaptive": [174], "mutation": [175], "ratio.": [179], "Simulation": [180], "analysis": [181], "demonstrates": [182], "that": [183], "effectively": [185], "prioritizes": [186], "high-priority": [190], "alleviate": [193], "competition,": [195], "achieving": [196], "highest": [198], "while": [202], "avoiding": [203], "excessive": [204], "network": [205], "reconfiguration.": [206]}, "primary_topic": {"id": "https://openalex.org/T10847", "display_name": "Advanced Optical Network Technologies", "score": 0.9824000000953674, "subfield": {"id": "https://openalex.org/subfields/2208", "display_name": "Electrical and Electronic Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10847", "display_name": "Advanced Optical Network Technologies", "score": 0.9824000000953674, "subfield": {"id": "https://openalex.org/subfields/2208", "display_name": "Electrical and Electronic Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10714", "display_name": "Software-Defined Networks and 5G", "score": 0.00930000003427267, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10138", "display_name": "Network Traffic and Congestion Control", "score": 0.003000000026077032, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/jiot.2025.3558933", "is_oa": false, "landing_page_url": "https://doi.org/10.1109/jiot.2025.3558933", "pdf_url": null, "source": {"id": "https://openalex.org/S2480266640", "display_name": "IEEE Internet of Things Journal", "issn_l": "2327-4662", "issn": ["2327-4662", "2372-2541"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-2", "norway-2"], "host_organization": "https://openalex.org/P4310319808", "host_organization_name": "Institute of Electrical and Electronics Engineers", "host_organization_lineage": ["https://openalex.org/P4310319808"], "host_organization_lineage_names": ["Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Internet of Things Journal", "raw_type": "journal-article"}, "cited_by_count": 3, "is_retracted": false}, {"id": "https://openalex.org/W4416011719", "doi": "https://doi.org/10.48550/arxiv.2505.03946", "title": "Decentralized Distributed Proximal Policy Optimization (DD-PPO) for High Performance Computing Scheduling on Multi-User Systems", "publication_year": 2025, "publication_date": "2025-05-06", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5012356742", "display_name": "Matthew Sgambati", "orcid": "https://orcid.org/0000-0002-9648-2045", "observed_orcids": ["https://orcid.org/0000-0002-9648-2045"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Sgambati, Matthew", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5005267824", "display_name": "Aleksandar Vakanski", "orcid": "https://orcid.org/0000-0003-3365-1291", "observed_orcids": ["https://orcid.org/0000-0003-3365-1291"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Vakanski, Aleksandar", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5011402768", "display_name": "Matthew Anderson", "orcid": "https://orcid.org/0000-0001-7997-6134", "observed_orcids": ["https://orcid.org/0000-0001-7997-6134"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Anderson, Matthew", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Resource": [0], "allocation": [1, 21], "in": [2, 72, 88, 236], "High": [3], "Performance": [4], "Computing": [5], "(HPC)": [6], "environments": [7], "presents": [8], "a": [9, 106, 133, 159, 194], "complex": [10], "and": [11, 29, 38, 56, 67, 77, 96, 119, 146, 204, 224, 228, 242], "multifaceted": [12], "challenge": [13, 64], "for": [14, 28, 218], "job": [15, 35, 74, 216], "scheduling": [16, 44, 98, 102, 226, 234, 245], "algorithms.": [17, 246], "Beyond": [18], "the": [19, 47, 53, 65, 164, 197, 229], "efficient": [20], "of": [22, 50, 58, 69, 108], "system": [23, 39], "resources,": [24], "schedulers": [25, 241], "must": [26], "account": [27], "optimize": [30], "multiple": [31, 178], "performance": [32, 221, 235], "metrics,": [33], "including": [34], "wait": [36, 75], "time": [37, 76], "utilization.": [40, 79, 206], "While": [41], "traditional": [42, 223], "rule-based": [43, 240], "algorithms": [45, 71], "dominate": [46], "current": [48], "deployments": [49], "HPC": [51, 215], "systems,": [52], "increasing": [54], "heterogeneity": [55], "scale": [57], "those": [59, 70], "systems": [60], "is": [61, 139], "expected": [62], "to": [63, 92, 114, 193, 238], "efficiency": [66], "flexibility": [68], "minimizing": [73], "maximizing": [78], "Recent": [80, 100], "research": [81], "efforts": [82], "have": [83, 104], "focused": [84], "on": [85, 142, 190], "leveraging": [86], "advancements": [87], "Reinforcement": [89], "Learning": [90], "(RL)": [91], "develop": [93], "more": [94, 120], "adaptable": [95], "intelligent": [97], "strategies.": [99], "RL-based": [101, 161, 244], "approaches": [103], "explored": [105], "range": [107], "algorithms,": [109], "from": [110], "Deep": [111], "Q-Networks": [112], "(DQN)": [113], "Proximal": [115, 167], "Policy": [116, 168], "Optimization": [117, 169], "(PPO),": [118], "recently,": [121], "hybrid": [122], "methods": [123, 138, 148], "that": [124], "integrate": [125], "Graph": [126], "Neural": [127], "Networks": [128], "with": [129], "RL": [130], "techniques.": [131], "However,": [132], "common": [134], "limitation": [135], "across": [136, 177], "these": [137, 147], "their": [140], "reliance": [141, 189], "relatively": [143], "small": [144], "datasets,": [145], "face": [149], "scalability": [150], "issues": [151], "when": [152], "using": [153], "large": [154], "datasets.": [155], "This": [156], "study": [157], "introduces": [158], "novel": [160], "scheduler": [162, 199], "utilizing": [163], "Decentralized": [165], "Distributed": [166], "(DD-PPO)": [170], "algorithm,": [171], "which": [172], "supports": [173], "large-scale": [174], "distributed": [175], "training": [176, 202], "workers": [179], "without": [180], "requiring": [181], "parameter": [182], "synchronization": [183], "at": [184], "every": [185], "step.": [186], "By": [187], "eliminating": [188], "centralized": [191], "updates": [192], "shared": [195], "policy,": [196], "DD-PPO": [198, 220], "enhances": [200], "scalability,": [201], "efficiency,": [203], "sample": [205], "The": [207], "validation": [208], "dataset": [209], "leveraged": [210], "over": [211], "11.5": [212], "million": [213], "real": [214], "traces": [217], "comparing": [219], "between": [222], "advanced": [225], "approaches,": [227], "experimental": [230], "results": [231], "demonstrate": [232], "improved": [233], "comparison": [237], "both": [239], "existing": [243]}, "primary_topic": {"id": "https://openalex.org/T10715", "display_name": "Distributed and Parallel Computing Systems", "score": 0.7523999810218811, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10715", "display_name": "Distributed and Parallel Computing Systems", "score": 0.7523999810218811, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.1096000000834465, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.04729999974370003, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2505.03946", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2505.03946", "pdf_url": "https://arxiv.org/pdf/2505.03946", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4407026180", "doi": "https://doi.org/10.1109/tgrs.2025.3537085", "title": "Toward Federated Customized Neural Architecture Search for Remote Sensing Scene Classification", "publication_year": 2025, "publication_date": "2025-01-01", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5103165266", "display_name": "Jianzhao Li", "orcid": "https://orcid.org/0000-0002-1524-1363", "observed_orcids": ["https://orcid.org/0000-0002-1524-1363"]}, "institutions": [{"id": "https://openalex.org/I149594827", "display_name": "Xidian University", "ror": "https://ror.org/05s92vm98", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I149594827"]}, {"id": "https://openalex.org/I4405280064", "display_name": "Guangzhou Institute of Technology", "ror": "https://ror.org/02d1g9y70", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I4405280064"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Jianzhao Li", "raw_affiliation_strings": ["Guangzhou Institute of Technology, Xidian University, Guangzhou, China"], "raw_orcid": "https://orcid.org/0000-0002-1524-1363", "affiliations": [{"raw_affiliation_string": "Guangzhou Institute of Technology, Xidian University, Guangzhou, China", "institution_ids": ["https://openalex.org/I149594827", "https://openalex.org/I4405280064"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5101427060", "display_name": "Shanfeng Wang", "orcid": "https://orcid.org/0000-0002-2151-6722", "observed_orcids": ["https://orcid.org/0000-0002-2151-6722"]}, "institutions": [{"id": "https://openalex.org/I149594827", "display_name": "Xidian University", "ror": "https://ror.org/05s92vm98", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I149594827"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Shanfeng Wang", "raw_affiliation_strings": ["Key Laboratory of Collaborative Intelligence Systems, Ministry of Education, Xidian University, Xi&#x2019;an, China"], "raw_orcid": "https://orcid.org/0000-0002-2151-6722", "affiliations": [{"raw_affiliation_string": "Key Laboratory of Collaborative Intelligence Systems, Ministry of Education, Xidian University, Xi&#x2019;an, China", "institution_ids": ["https://openalex.org/I149594827"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5115591390", "display_name": "Rui Yang", "orcid": "https://orcid.org/0000-0001-5495-0165", "observed_orcids": ["https://orcid.org/0000-0001-5495-0165"]}, "institutions": [{"id": "https://openalex.org/I149594827", "display_name": "Xidian University", "ror": "https://ror.org/05s92vm98", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I149594827"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Rui Yang", "raw_affiliation_strings": ["Key Laboratory of Collaborative Intelligence Systems, Ministry of Education, Xidian University, Xi&#x2019;an, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Key Laboratory of Collaborative Intelligence Systems, Ministry of Education, Xidian University, Xi&#x2019;an, China", "institution_ids": ["https://openalex.org/I149594827"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5091227928", "display_name": "Maoguo Gong", "orcid": "https://orcid.org/0000-0002-0415-8556", "observed_orcids": ["https://orcid.org/0000-0002-0415-8556"]}, "institutions": [{"id": "https://openalex.org/I149594827", "display_name": "Xidian University", "ror": "https://ror.org/05s92vm98", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I149594827"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Maoguo Gong", "raw_affiliation_strings": ["School of Cyber Engineering, Key Laboratory of Collaborative Intelligence Systems, Ministry of Education, Xidian University, Xi&#x2019;an, China"], "raw_orcid": "https://orcid.org/0000-0002-0415-8556", "affiliations": [{"raw_affiliation_string": "School of Cyber Engineering, Key Laboratory of Collaborative Intelligence Systems, Ministry of Education, Xidian University, Xi&#x2019;an, China", "institution_ids": ["https://openalex.org/I149594827"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5024283292", "display_name": "Zhuping Hu", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I22046295", "display_name": "Inner Mongolia Normal University", "ror": "https://ror.org/0497ase59", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I22046295"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Zhuping Hu", "raw_affiliation_strings": ["Academy of Artificial Intelligence, College of Mathematics Science, Inner Mongolia Normal University, Hohhot, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Academy of Artificial Intelligence, College of Mathematics Science, Inner Mongolia Normal University, Hohhot, China", "institution_ids": ["https://openalex.org/I22046295"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100404886", "display_name": "Ning Zhang", "orcid": "https://orcid.org/0000-0002-8781-4925", "observed_orcids": ["https://orcid.org/0000-0002-8781-4925", "https://orcid.org/0000-0001-9005-937X", "https://orcid.org/0009-0004-5887-2327"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Ning Zhang", "raw_affiliation_strings": ["Shanghai Aerospace Electronic Technology Institute, Shanghai, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Shanghai Aerospace Electronic Technology Institute, Shanghai, China", "institution_ids": []}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5056601955", "display_name": "Kai Sheng", "orcid": "https://orcid.org/0000-0001-6602-7212", "observed_orcids": ["https://orcid.org/0000-0001-6602-7212"]}, "institutions": [{"id": "https://openalex.org/I149594827", "display_name": "Xidian University", "ror": "https://ror.org/05s92vm98", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I149594827"]}, {"id": "https://openalex.org/I4405280064", "display_name": "Guangzhou Institute of Technology", "ror": "https://ror.org/02d1g9y70", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I4405280064"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Kai Sheng", "raw_affiliation_strings": ["Guangzhou Institute of Technology, Xidian University, Guangzhou, China"], "raw_orcid": "https://orcid.org/0000-0001-6602-7212", "affiliations": [{"raw_affiliation_string": "Guangzhou Institute of Technology, Xidian University, Guangzhou, China", "institution_ids": ["https://openalex.org/I149594827", "https://openalex.org/I4405280064"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5024112348", "display_name": "Yu Zhou", "orcid": "https://orcid.org/0009-0006-9950-4728", "observed_orcids": ["https://orcid.org/0009-0006-9950-4728"]}, "institutions": [{"id": "https://openalex.org/I22046295", "display_name": "Inner Mongolia Normal University", "ror": "https://ror.org/0497ase59", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I22046295"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yu Zhou", "raw_affiliation_strings": ["Academy of Artificial Intelligence, College of Mathematics Science, Inner Mongolia Normal University, Hohhot, China"], "raw_orcid": "https://orcid.org/0009-0006-9950-4728", "affiliations": [{"raw_affiliation_string": "Academy of Artificial Intelligence, College of Mathematics Science, Inner Mongolia Normal University, Hohhot, China", "institution_ids": ["https://openalex.org/I22046295"]}]}], "abstract_inverted_index": {"Remote": [0], "sensing": [1, 75, 128], "(RS)": [2], "scenarios": [3], "usually": [4], "involve": [5], "sensitive": [6], "geographic": [7], "information": [8], "on": [9, 31, 203], "national": [10], "security": [11, 43], "and": [12, 29, 41, 80, 98, 195, 200, 211, 216], "regional": [13], "development.": [14], "In": [15, 67], "the": [16, 51, 133, 169, 176, 209, 217], "commonly": [17], "used": [18], "centralized": [19], "machine-learning": [20], "paradigm,": [21], "data": [22, 42, 145, 191, 193], "dispersed": [23], "in": [24, 114, 142, 166, 188], "various": [25], "locations": [26], "are": [27], "concentrated": [28], "processed": [30], "a": [32, 72, 82, 149, 156, 180], "single": [33], "server,": [34], "which": [35], "is": [36, 47, 109, 164], "prone": [37], "to": [38, 49, 64, 88, 111, 126, 147, 224], "privacy": [39, 146], "leakage": [40], "concerns.": [44], "Besides,": [45], "it": [46], "difficult": [48], "solve": [50], "high": [52], "heterogeneity": [53], "of": [54, 171, 190, 213], "RS": [55, 103, 173], "images": [56], "by": [57], "simply": [58], "applying": [59], "federated": [60, 73, 182], "learning": [61], "(FL)": [62], "algorithms": [63], "scene": [65, 76, 104, 129], "classification.": [66], "this": [68], "article,": [69], "we": [70], "formulate": [71], "remote": [74, 127], "classification": [77, 130], "(FedSC)": [78], "framework,": [79], "design": [81], "customized": [83, 157], "neural": [84], "architecture": [85], "search": [86], "(CNAS)": [87], "achieve": [89], "both": [90], "global": [91, 151], "generality": [92], "for": [93, 101, 160, 186], "multiparty": [94], "collaborative": [95, 139, 183], "distributed": [96, 140], "training": [97, 141, 184], "local": [99, 158], "specificity": [100], "personalized": [102, 196], "customization.": [105, 197], "The": [106], "proposed": [107, 177, 218], "FedSC": [108], "generalizable": [110], "be": [112], "implemented": [113], "any": [115], "manually": [116], "designed": [117, 134], "networks,": [118], "network": [119], "pruning": [120], "strategies,": [121], "or": [122], "NAS": [123], "methods": [124], "related": [125], "(RSSC).": [131], "While": [132], "CNAS": [135], "not": [136], "only": [137], "achieves": [138], "protecting": [143], "participant": [144, 162], "obtain": [148], "generalized": [150], "model,": [152], "but": [153], "also": [154], "provides": [155], "model": [159], "each": [161], "that": [163], "more": [165], "line": [167], "with": [168], "characteristics": [170], "private": [172], "scenarios.": [174], "Overall,": [175], "FedSC$_{\\textrm": [178, 219], "{CNAS}}$provides": [179], "novel": [181], "paradigm": [185], "RSSC": [187, 206], "terms": [189], "privacy,": [192], "heterogeneity,": [194], "Extensive": [198], "analytical": [199], "comparative": [201], "experiments": [202], "three": [204], "benchmark": [205], "datasets": [207], "validate": [208], "versatility": [210], "effectiveness": [212], "our": [214], "methods,": [215], "{CNAS}}$exhibits": [220], "superior": [221], "competitiveness": [222], "compared": [223], "state-of-the-art": [225], "methods.": [226]}, "primary_topic": {"id": "https://openalex.org/T10689", "display_name": "Remote-Sensing Image Classification", "score": 0.7020999789237976, "subfield": {"id": "https://openalex.org/subfields/2214", "display_name": "Media Technology"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10689", "display_name": "Remote-Sensing Image Classification", "score": 0.7020999789237976, "subfield": {"id": "https://openalex.org/subfields/2214", "display_name": "Media Technology"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.263700008392334, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.009600000455975533, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/tgrs.2025.3537085", "is_oa": false, "landing_page_url": "https://doi.org/10.1109/tgrs.2025.3537085", "pdf_url": null, "source": {"id": "https://openalex.org/S111326731", "display_name": "IEEE Transactions on Geoscience and Remote Sensing", "issn_l": "0196-2892", "issn": ["0196-2892", "1558-0644"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-3", "norway-2"], "host_organization": "https://openalex.org/P4310319808", "host_organization_name": "Institute of Electrical and Electronics Engineers", "host_organization_lineage": ["https://openalex.org/P4310319808"], "host_organization_lineage_names": ["Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Transactions on Geoscience and Remote Sensing", "raw_type": "journal-article"}, "cited_by_count": 4, "is_retracted": false}, {"id": "https://openalex.org/W4407699442", "doi": "https://doi.org/10.1103/physrevresearch.7.013177", "title": "Quantum data parallelism in quantum neural networks", "publication_year": 2025, "publication_date": "2025-02-18", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5110778369", "display_name": "Sixuan Wu", "orcid": "https://orcid.org/0009-0004-9377-4511", "observed_orcids": ["https://orcid.org/0009-0004-9377-4511"]}, "institutions": [{"id": "https://openalex.org/I24943067", "display_name": "Fudan University", "ror": "https://ror.org/013q1eq08", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I24943067"]}, {"id": "https://openalex.org/I3133055985", "display_name": "Westlake University", "ror": "https://ror.org/05hfa4n20", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I3133055985"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Sixuan Wu", "raw_affiliation_strings": ["Fudan University", "Westlake University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Fudan University", "institution_ids": ["https://openalex.org/I24943067"]}, {"raw_affiliation_string": "Westlake University", "institution_ids": ["https://openalex.org/I3133055985"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100333729", "display_name": "Yue Zhang", "orcid": "https://orcid.org/0000-0002-5214-2268", "observed_orcids": ["https://orcid.org/0000-0002-5214-2268"]}, "institutions": [{"id": "https://openalex.org/I3133055985", "display_name": "Westlake University", "ror": "https://ror.org/05hfa4n20", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I3133055985"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yue Zhang", "raw_affiliation_strings": ["Westlake University"], "raw_orcid": "https://orcid.org/0000-0002-5214-2268", "affiliations": [{"raw_affiliation_string": "Westlake University", "institution_ids": ["https://openalex.org/I3133055985"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5010229002", "display_name": "Jian Li", "orcid": "https://orcid.org/0000-0003-0297-6528", "observed_orcids": ["https://orcid.org/0000-0003-0297-6528"]}, "institutions": [{"id": "https://openalex.org/I3133055985", "display_name": "Westlake University", "ror": "https://ror.org/05hfa4n20", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I3133055985"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Jian Li", "raw_affiliation_strings": ["Westlake University"], "raw_orcid": "https://orcid.org/0000-0003-0297-6528", "affiliations": [{"raw_affiliation_string": "Westlake University", "institution_ids": ["https://openalex.org/I3133055985"]}]}], "abstract_inverted_index": {"Quantum": [0], "neural": [1, 35, 45, 68, 145], "networks": [2, 146], "hold": [3], "promise": [4], "for": [5, 78], "achieving": [6], "lower": [7], "generalization": [8], "error": [9], "bounds": [10], "and": [11, 59, 84, 97, 125], "enhanced": [12], "computational": [13], "efficiency": [14], "in": [15, 25, 30, 40, 65, 102, 138], "processing": [16], "certain": [17], "datasets.": [18], "However,": [19], "the": [20, 31, 41, 50, 79, 105, 154], "integration": [21], "of": [22, 33, 43, 53, 75, 104, 156], "quantum": [23, 44, 54, 57, 67, 86, 113, 126, 144, 157], "superposition": [24, 58], "data": [26, 63, 129, 151, 158], "parallelism\u2014a": [27], "key": [28], "aspect": [29], "applications": [32], "classical": [34], "network\u2014has": [36], "been": [37], "largely": [38], "unexplored": [39], "context": [42], "networks.": [46], "Here,": [47], "we": [48, 92, 118], "demonstrate": [49], "effective": [51], "application": [52], "parallelism,": [55], "via": [56], "entanglement,": [60], "to": [61], "achieve": [62], "parallelism": [64], "generic": [66], "network": [69], "models.": [70], "We": [71, 132], "address": [72], "two": [73], "classes": [74], "encoding": [76], "schemes": [77], "training": [80, 99], "data:": [81], "as": [82], "orthogonal": [83, 90], "nonorthogonal": [85, 116], "states,": [87], "respectively.": [88], "With": [89], "encoding,": [91, 117], "rigorously": [93], "prove": [94], "that": [95, 120], "parallel": [96, 121], "individual": [98], "are": [100], "equivalent": [101], "terms": [103], "loss": [106], "function,": [107], "a": [108, 134], "result": [109], "directly": [110], "derived": [111], "from": [112], "entanglement;": [114], "with": [115, 147], "show": [119], "computing": [122], "remains": [123], "viable,": [124], "interference": [127], "among": [128], "becomes": [130], "notable.": [131], "perform": [133], "comparative": [135], "numerical": [136], "analysis": [137], "these": [139], "different": [140], "scenarios,": [141], "exploring": [142], "various": [143], "or": [148], "without": [149], "superposed": [150], "input,": [152], "highlighting": [153], "validity": [155], "parallelism.": [159]}, "primary_topic": {"id": "https://openalex.org/T10682", "display_name": "Quantum Computing Algorithms and Architecture", "score": 1.0, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10682", "display_name": "Quantum Computing Algorithms and Architecture", "score": 1.0, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11804", "display_name": "Quantum many-body systems", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/3107", "display_name": "Atomic and Molecular Physics, and Optics"}, "field": {"id": "https://openalex.org/fields/31", "display_name": "Physics and Astronomy"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10020", "display_name": "Quantum Information and Cryptography", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1103/physrevresearch.7.013177", "is_oa": true, "landing_page_url": "https://doi.org/10.1103/physrevresearch.7.013177", "pdf_url": null, "source": {"id": "https://openalex.org/S4210240247", "display_name": "Physical Review Research", "issn_l": "2643-1564", "issn": ["2643-1564"], "is_oa": true, "is_in_doaj": true, "is_core": true, "listed_in": ["cwts-core", "doaj", "jufo-1", "norway-1"], "host_organization": "https://openalex.org/P4310320261", "host_organization_name": "American Physical Society", "host_organization_lineage": ["https://openalex.org/P4310320261"], "host_organization_lineage_names": ["American Physical Society"], "type": "journal"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Physical Review Research", "raw_type": "journal-article"}, "cited_by_count": 11, "is_retracted": false}, {"id": "https://openalex.org/W4417071098", "doi": "https://doi.org/10.48550/arxiv.2509.09009", "title": "Open-sci-ref-0.01: open and reproducible reference baselines for language model and dataset comparison", "publication_year": 2025, "publication_date": "2025-09-10", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5042252873", "display_name": "Marianna Nezhurina", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Nezhurina, Marianna", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5059197069", "display_name": "J\u00f6rg E. Franke", "orcid": "https://orcid.org/0000-0003-0700-2028", "observed_orcids": ["https://orcid.org/0000-0003-0700-2028", "https://orcid.org/0000-0001-7897-4225"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Franke, J\u00f6rg", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5039154090", "display_name": "Taishi Nakamura", "orcid": "https://orcid.org/0000-0002-1205-3252", "observed_orcids": ["https://orcid.org/0000-0002-1205-3252"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Nakamura, Taishi", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5116293279", "display_name": "Timur Carstensen", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Carstensen, Timur", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5114337674", "display_name": "Niccol\u00f2 Ajroldi", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Ajroldi, Niccol\u00f2", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5003808414", "display_name": "Ville Komulainen", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Komulainen, Ville", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5078696011", "display_name": "David Salinas", "orcid": "https://orcid.org/0000-0002-8980-4018", "observed_orcids": ["https://orcid.org/0000-0002-8980-4018"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Salinas, David", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5085870953", "display_name": "Jenia Jitsev", "orcid": "https://orcid.org/0000-0002-1221-7851", "observed_orcids": ["https://orcid.org/0000-0002-1221-7851"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Jitsev, Jenia", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"We": [0], "introduce": [1], "open-sci-ref,": [2], "a": [3, 90], "family": [4], "of": [5, 55, 69, 95], "dense": [6], "transformer": [7], "models": [8, 34], "trained": [9], "as": [10], "research": [11], "baselines": [12, 76], "across": [13, 59], "multiple": [14], "model": [15], "(0.13B": [16], "to": [17, 24, 49, 80, 117, 129], "1.7B": [18], "parameters)": [19], "and": [20, 53, 61, 67, 113, 126, 134], "token": [21], "scales": [22, 60], "(up": [23], "1T)": [25], "on": [26, 35, 89, 102], "8": [27], "recent": [28], "open": [29, 96], "reference": [30, 44, 75, 97, 108], "datasets.": [31, 62], "Evaluating": [32], "the": [33, 51, 70, 121], "various": [36], "standardized": [37], "benchmarks,": [38], "our": [39], "training": [40, 57, 71, 78, 101, 119], "runs": [41], "set": [42], "establishes": [43], "points": [45], "that": [46, 100], "enable": [47], "researchers": [48], "assess": [50], "sanity": [52], "quality": [54], "alternative": [56], "approaches": [58], "Intermediate": [63], "checkpoints": [64], "allow": [65, 77], "comparison": [66], "studying": [68], "dynamics.": [72], "The": [73], "established": [74], "procedures": [79], "be": [81], "compared": [82], "through": [83], "their": [84], "scaling": [85], "trends,": [86], "aligning": [87], "them": [88], "common": [91], "compute": [92], "axis.": [93], "Comparison": [94], "datasets": [98], "reveals": [99], "NemoTron-CC": [103], "HQ": [104], "consistently": [105], "outperforms": [106], "other": [107], "datasets,": [109], "followed": [110], "by": [111], "DCLM-baseline": [112], "FineWeb-Edu.": [114], "In": [115], "addition": [116], "intermediate": [118], "checkpoints,": [120], "release": [122], "includes": [123], "logs,": [124], "code,": [125], "downstream": [127], "evaluations": [128], "simplify": [130], "reproduction,": [131], "standardize": [132], "comparison,": [133], "facilitate": [135], "future": [136], "research.": [137]}, "primary_topic": {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.9413999915122986, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.9413999915122986, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10181", "display_name": "Natural Language Processing Techniques", "score": 0.048700001090765, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11307", "display_name": "Domain Adaptation and Few-Shot Learning", "score": 0.0013000000035390258, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2509.09009", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2509.09009", "pdf_url": "https://arxiv.org/pdf/2509.09009", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4417101148", "doi": "https://doi.org/10.48550/arxiv.2508.01840", "title": "Implementing Neural Networks Over-the-Air via Reconfigurable Intelligent Surfaces", "publication_year": 2025, "publication_date": "2025-08-03", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Hua, Meng", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Hua, Meng", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5076857062", "display_name": "Chenghong Bian", "orcid": "https://orcid.org/0000-0002-0534-7076", "observed_orcids": ["https://orcid.org/0000-0002-0534-7076"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Bian, Chenghong", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100709658", "display_name": "Haotian Wu", "orcid": "https://orcid.org/0000-0001-5204-4388", "observed_orcids": ["https://orcid.org/0000-0001-5204-4388"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Wu, Haotian", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5016883501", "display_name": "Den\u0131z G\u00fcnd\u00fcz", "orcid": "https://orcid.org/0000-0002-7725-395X", "observed_orcids": ["https://orcid.org/0000-0002-7725-395X"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Gunduz, Deniz", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"In": [0, 134, 166], "this": [1, 56, 95], "paper,": [2], "we": [3, 99, 116, 198], "investigate": [4], "reconfigurable": [5], "intelligent": [6], "surface": [7], "(RIS)-aided": [8], "multiple-input-multiple-output": [9], "(MIMO)": [10], "OAC": [11, 86], "systems": [12], "designed": [13], "to": [14, 37, 44, 55, 80, 162, 202, 212], "emulate": [15, 45], "the": [16, 29, 32, 39, 46, 49, 65, 69, 82, 85, 89, 120, 135, 144, 147, 151, 157, 163, 167, 171, 214, 221, 226], "fully-connected": [17], "(FC)": [18], "layer": [19], "of": [20, 48, 119], "a": [21, 101, 203], "neural": [22], "network": [23], "(NN)": [24], "via": [25], "analog": [26], "OAC,": [27], "where": [28, 106], "RIS": [30, 73], "and": [31, 72, 88, 131, 156, 173, 181, 192], "transceivers": [33], "are": [34, 77, 113, 160], "jointly": [35, 78], "adjusted": [36], "engineer": [38], "ambient": [40], "wireless": [41], "propagation": [42], "environment": [43], "weights": [47], "target": [50, 90], "FC": [51, 91], "layer.": [52, 92], "We": [53, 62], "refer": [54], "novel": [57], "computational": [58, 195], "paradigm": [59], "as": [60], "AirFC.": [61], "first": [63], "study": [64], "case": [66], "in": [67], "which": [68], "precoder,": [70], "combiner,": [71], "phase": [74], "shift": [75], "matrices": [76], "optimized": [79], "minimize": [81], "mismatch": [83], "between": [84], "system": [87, 121, 215, 223], "To": [93], "solve": [94], "non-convex": [96], "optimization": [97, 104, 111], "problem,": [98], "propose": [100], "low-complexity": [102], "alternating": [103], "algorithm,": [105], "semi-closed-form/closed-form": [107], "solutions": [108], "for": [109], "all": [110], "variables": [112], "derived.": [114], "Next,": [115], "consider": [117], "training": [118, 130, 137, 139, 169], "parameters": [122, 159, 178], "using": [123], "two": [124], "distinct": [125], "learning": [126], "strategies,": [127], "namely": [128], "centralized": [129, 136], "distributed": [132, 168], "training.": [133], "approach,": [138, 170], "is": [140], "performed": [141], "at": [142], "either": [143], "transmitter": [145, 172], "or": [146], "receiver,": [148], "whichever": [149], "possesses": [150], "channel": [152, 186], "state": [153], "information": [154], "(CSI),": [155], "trained": [158], "provided": [161], "other": [164], "terminal.": [165], "receiver": [174], "iteratively": [175], "update": [176], "their": [177], "through": [179], "back": [180], "forth": [182], "transmissions": [183], "by": [184, 206, 225], "leveraging": [185], "reciprocity,": [187], "thereby": [188], "avoiding": [189], "CSI": [190], "acquisition": [191], "significantly": [193], "reducing": [194], "complexity.": [196], "Subsequently,": [197], "extend": [199], "our": [200], "analysis": [201], "multi-RIS": [204], "scenario": [205], "exploiting": [207], "its": [208], "spatial": [209], "diversity": [210], "gain": [211], "enhance": [213], "performance.": [216], "Simulation": [217], "results": [218], "show": [219], "that": [220], "AirFC": [222], "realized": [224], "RIS-aided": [227], "MIMO": [228], "configuration": [229], "achieves": [230], "satisfactory": [231], "classification": [232], "accuracy.": [233]}, "primary_topic": {"id": "https://openalex.org/T11458", "display_name": "Advanced Wireless Communication Technologies", "score": 0.9994999766349792, "subfield": {"id": "https://openalex.org/subfields/2208", "display_name": "Electrical and Electronic Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T11458", "display_name": "Advanced Wireless Communication Technologies", "score": 0.9994999766349792, "subfield": {"id": "https://openalex.org/subfields/2208", "display_name": "Electrical and Electronic Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11383", "display_name": "Advanced Antenna and Metasurface Technologies", "score": 9.999999747378752e-05, "subfield": {"id": "https://openalex.org/subfields/2202", "display_name": "Aerospace Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10851", "display_name": "Optical Wireless Communication Technologies", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/2208", "display_name": "Electrical and Electronic Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2508.01840", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2508.01840", "pdf_url": "https://arxiv.org/pdf/2508.01840", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W7219326927", "doi": null, "title": "Enhancing Large-Scale Medical Image Processing in HPC Using Distributed Convolutional Neural Networks", "publication_year": 2025, "publication_date": "2025-05-08", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5073946656", "display_name": "Abhishek Gandhar", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Gandhar, Abhishek", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5156688249", "display_name": "S. B. Kumar", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Kumar, S. B.", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5028006222", "display_name": "Shashi Gandhar", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Gandhar, Shashi", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5086755066", "display_name": "Prakhar Priyadarshi", "orcid": "https://orcid.org/0000-0003-0367-840X", "observed_orcids": ["https://orcid.org/0000-0003-0367-840X"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Priyadarshi, Prakhar", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5050501207", "display_name": "Arvind Rehalia", "orcid": "https://orcid.org/0000-0002-2184-3432", "observed_orcids": ["https://orcid.org/0000-0002-2184-3432"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Rehalia,  Arvind", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5022861299", "display_name": "Mohit Tiwari", "orcid": "https://orcid.org/0000-0003-1836-3451", "observed_orcids": ["https://orcid.org/0000-0003-1836-3451"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Tiwari, Mohit", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"The": [0, 42], "demand": [1], "for": [2, 30, 85, 195], "efficient": [3, 193], "and": [4, 49, 59, 61, 103, 134, 171, 176, 192, 204], "scalable": [5, 191], "processing": [6, 40, 162], "techniques": [7], "in": [8, 92, 96, 174, 200], "the": [9, 68, 86, 97, 121, 124, 137, 147, 154, 183, 215], "era": [10], "of": [11, 77, 99, 108, 157, 185, 202], "high-resolution": [12], "medical": [13, 38, 139, 196], "imaging": [14, 140], "has": [15], "grown": [16], "significantly.": [17], "This": [18], "research": [19], "introduces": [20], "a": [21, 72, 100, 109, 112, 190], "distributed": [22], "Convolutional": [23], "type": [24], "Neural-based": [25], "Network": [26], "(CNN)-based": [27], "framework": [28], "optimized": [29], "Higher-Performance": [31], "Computation": [32], "(HPC)": [33], "environments": [34], "to": [35, 53, 119, 131], "perform": [36, 120], "large-scale": [37], "image": [39, 197], "better.": [41], "proposed": [43], "system": [44, 165], "utilizes": [45], "parallel": [46, 94], "computing": [47], "architectures": [48], "data": [50, 78, 163], "partitioning": [51], "strategies": [52], "accelerate": [54], "deep": [55, 87, 218], "learning-based": [56], "feature": [57], "extraction": [58], "classification": [60], "executes": [62], "multiple": [63], "training": [64], "tasks": [65], "simultaneously": [66], "on": [67], "different": [69, 138], "nodes": [70, 116], "within": [71], "cluster.": [73], "A": [74, 160], "two-step": [75], "approach": [76], "parallelism": [79, 82], "with": [80, 114], "model": [81], "is": [83, 128, 144, 212], "adopted": [84], "neural": [88, 101], "network": [89, 102, 110], "training,": [90], "which": [91, 152, 199], "turn": [93], "manner": [95], "layers": [98], "breaks": [104], "down": [105], "each": [106], "layer": [107], "into": [111], "sub-network,": [113], "all": [115], "working": [117], "together": [118], "computation.": [122], "Furthermore,": [123], "adaptive": [125], "learning": [126, 219], "mechanism": [127], "also": [129], "embedded": [130], "improve": [132], "convergence": [133], "generalization": [135], "across": [136], "modalities.": [141], "Further": [142], "agility": [143], "ensured": [145], "by": [146], "dynamic": [148], "workload": [149], "scheduling": [150], "strategy": [151], "guarantees": [153], "effective": [155], "distribution": [156], "computational": [158, 205], "schools.": [159], "real-time": [161], "analysis": [164], "will": [166], "help": [167], "diagnose": [168], "patients": [169], "quickly": [170], "accurately,": [172], "resulting": [173], "better": [175], "faster": [177], "clinical": [178], "decision": [179], "making.": [180], "By": [181], "describing": [182], "capabilities": [184], "HPC,": [186], "this": [187], "method": [188], "presents": [189], "solution": [194], "analysis,": [198], "terms": [201], "speed": [203], "efficiency": [206], "provides": [207], "significant": [208], "improvements": [209], "as": [210], "it": [211], "offered": [213], "via": [214], "traditional": [216], "centralized": [217], "methods.": [220]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.37400001287460327, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.37400001287460327, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.31349998712539673, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11636", "display_name": "Artificial Intelligence in Healthcare and Education", "score": 0.06920000165700912, "subfield": {"id": "https://openalex.org/subfields/2718", "display_name": "Health Informatics"}, "field": {"id": "https://openalex.org/fields/27", "display_name": "Medicine"}, "domain": {"id": "https://openalex.org/domains/4", "display_name": "Health Sciences"}}], "primary_location": {"id": "pmh:oai:jneonatalsurg.com:article/5348", "is_oa": true, "landing_page_url": "https://www.jneonatalsurg.com/index.php/jns/article/view/5348", "pdf_url": null, "source": {"id": "https://openalex.org/S2764549079", "display_name": "Journal of Neonatal Surgery", "issn_l": "2226-0439", "issn": ["2226-0439"], "is_oa": true, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core"], "host_organization": null, "host_organization_name": null, "host_organization_lineage": [], "host_organization_lineage_names": [], "type": "journal"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "2226-0439", "raw_type": "info:eu-repo/semantics/article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4417068969", "doi": "https://doi.org/10.48550/arxiv.2503.23186", "title": "Optimizing Distributed Training Approaches for Scaling Neural Networks", "publication_year": 2025, "publication_date": "2025-03-29", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5038960912", "display_name": "Vishnu Vardhan Baligodugula", "orcid": "https://orcid.org/0000-0003-2223-4396", "observed_orcids": ["https://orcid.org/0000-0003-2223-4396"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Baligodugula, Vishnu Vardhan", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5045929043", "display_name": "Fathi Amsaad", "orcid": "https://orcid.org/0000-0002-1641-5046", "observed_orcids": ["https://orcid.org/0000-0002-1641-5046", "https://orcid.org/0000-0002-7582-8326"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Amsaad, Fathi", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"This": [0], "paper": [1], "presents": [2], "a": [3, 51], "comparative": [4], "analysis": [5], "of": [6], "distributed": [7], "training": [8, 36, 57, 89], "strategies": [9, 26, 73], "for": [10], "large-scale": [11], "neural": [12], "networks,": [13], "focusing": [14], "on": [15, 27, 75], "data": [16], "parallelism,": [17, 19], "model": [18, 41], "and": [20, 40, 78], "hybrid": [21, 48], "approaches.": [22], "We": [23, 62], "evaluate": [24], "these": [25], "image": [28], "classification": [29], "tasks": [30], "using": [31], "the": [32], "CIFAR-100": [33], "dataset,": [34], "measuring": [35], "time,": [37], "convergence": [38], "rate,": [39], "accuracy.": [42, 61], "Our": [43], "experimental": [44], "results": [45], "demonstrate": [46], "that": [47, 68], "parallelism": [49, 72], "achieves": [50], "3.2x": [52], "speedup": [53], "compared": [54], "to": [55], "single-device": [56], "while": [58], "maintaining": [59], "comparable": [60], "propose": [63], "an": [64, 84], "adaptive": [65], "scheduling": [66], "algorithm": [67], "dynamically": [69], "switches": [70], "between": [71], "based": [74], "network": [76], "characteristics": [77], "available": [79], "computational": [80], "resources,": [81], "resulting": [82], "in": [83, 88], "additional": [85], "18%": [86], "improvement": [87], "efficiency.": [90]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.5464000105857849, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.5464000105857849, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.28439998626708984, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.062199998646974564, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2503.23186", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2503.23186", "pdf_url": "https://arxiv.org/pdf/2503.23186", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4415935366", "doi": "https://doi.org/10.48550/arxiv.2511.00055", "title": "Exploring Federated Learning for Thermal Urban Feature Segmentation -- A Comparison of Centralized and Decentralized Approaches", "publication_year": 2025, "publication_date": "2025-10-28", "type": "preprint", "language": null, "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5118678215", "display_name": "Leonhard Duda", "orcid": "https://orcid.org/0009-0008-2432-9392", "observed_orcids": ["https://orcid.org/0009-0008-2432-9392"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Duda, Leonhard", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5028642208", "display_name": "Khadijeh Alibabaei", "orcid": "https://orcid.org/0000-0002-2319-8211", "observed_orcids": ["https://orcid.org/0000-0002-2319-8211"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Alibabaei, Khadijeh", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5022691676", "display_name": "Elena Vollmer", "orcid": "https://orcid.org/0000-0002-8805-3726", "observed_orcids": ["https://orcid.org/0000-0002-8805-3726"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Vollmer, Elena", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5116085628", "display_name": "Leon Klug", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Klug, Leon", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5103105872", "display_name": "Valentin Y. Kozlov", "orcid": "https://orcid.org/0000-0002-8770-3619", "observed_orcids": ["https://orcid.org/0000-0002-8770-3619"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Kozlov, Valentin", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5066919424", "display_name": "Lisana Berberi", "orcid": "https://orcid.org/0000-0002-7632-2466", "observed_orcids": ["https://orcid.org/0000-0002-7632-2466"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Berberi, Lisana", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5048011174", "display_name": "Mishal Benz", "orcid": "https://orcid.org/0009-0006-1491-7567", "observed_orcids": ["https://orcid.org/0009-0006-1491-7567"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Benz, Mishal", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5028481845", "display_name": "Rebekka Volk", "orcid": "https://orcid.org/0000-0001-9930-5354", "observed_orcids": ["https://orcid.org/0000-0001-9930-5354"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Volk, Rebekka", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5092682913", "display_name": "Juan Pedro Guti\u00e9rrez Hermosillo Muriedas", "orcid": "https://orcid.org/0000-0001-8439-7145", "observed_orcids": ["https://orcid.org/0000-0001-8439-7145"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Muriedas, Juan Pedro Guti\u00e9rrez Hermosillo", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5067996051", "display_name": "Markus G\u00f6tz", "orcid": "https://orcid.org/0000-0002-2233-1041", "observed_orcids": ["https://orcid.org/0000-0002-2233-1041"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "G\u00f6tz, Markus", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5002652575", "display_name": "Judith S\u00e1inz-Pardo D\u00edaz", "orcid": "https://orcid.org/0000-0002-8387-578X", "observed_orcids": ["https://orcid.org/0000-0002-8387-578X"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "D\u00edaz, Judith S\u00e1\u00ednz-Pardo", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5072424728", "display_name": "\u00c1lvaro L\u00f3pez Garc\u00eda", "orcid": "https://orcid.org/0000-0002-0013-4602", "observed_orcids": ["https://orcid.org/0000-0002-0013-4602"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Garc\u00eda, \u00c1lvaro L\u00f3pez", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5051993448", "display_name": "Frank Schultmann", "orcid": "https://orcid.org/0000-0001-6405-9763", "observed_orcids": ["https://orcid.org/0000-0001-6405-9763"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Schultmann, Frank", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5030034618", "display_name": "Achim Streit", "orcid": "https://orcid.org/0000-0002-5065-469X", "observed_orcids": ["https://orcid.org/0000-0002-5065-469X"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Streit, Achim", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Federated": [0], "Learning": [1, 11, 30], "(FL)": [2], "is": [3], "an": [4], "approach": [5], "for": [6, 91, 111, 206], "training": [7, 16, 55, 176], "a": [8, 79, 164, 203], "shared": [9, 36], "Machine": [10, 29], "(ML)": [12], "model": [13, 51, 174], "with": [14, 53, 163], "distributed": [15, 100], "data": [17, 33, 56, 104, 136], "and": [18, 57, 74, 107, 132, 180, 193, 211], "multiple": [19], "participants.": [20, 67], "FL": [21, 77, 112, 149, 161, 188, 215], "allows": [22], "bypassing": [23], "limitations": [24, 212], "of": [25, 76, 102, 135, 198, 213], "the": [26, 47, 50, 65, 71, 103, 208, 214], "traditional": [27], "Centralized": [28], "CL": [31], "if": [32], "cannot": [34], "be": [35], "or": [37, 43], "stored": [38], "centrally": [39], "due": [40, 128], "to": [41, 61, 129], "privacy": [42], "technical": [44], "restrictions": [45], "--": [46], "participants": [48], "train": [49], "locally": [52], "their": [54], "do": [58], "not": [59], "need": [60], "share": [62], "it": [63, 109], "among": [64], "other": [66], "This": [68, 123, 183], "paper": [69, 184], "investigates": [70], "practical": [72, 209], "implementation": [73], "effectiveness": [75], "in": [78, 96, 117, 151, 217, 220], "real-world": [80], "scenario,": [81], "specifically": [82], "focusing": [83], "on": [84], "unmanned": [85], "aerial": [86], "vehicle": [87], "(UAV)-based": [88], "thermal": [89, 93], "images": [90, 115], "common": [92], "feature": [94, 133], "detection": [95], "urban": [97], "environments.": [98], "The": [99, 141, 196], "nature": [101], "arises": [105], "naturally": [106], "makes": [108, 143], "suitable": [110], "applications,": [113], "as": [114, 173, 202], "captured": [116, 137], "two": [118], "German": [119], "cities": [120], "are": [121], "available.": [122], "application": [124, 210], "presents": [125], "unique": [126], "challenges": [127], "non-identical": [130], "distribution": [131], "characteristics": [134], "at": [138], "both": [139], "locations.": [140], "study": [142], "several": [144, 160], "key": [145, 169], "contributions": [146], "by": [147], "evaluating": [148], "algorithms": [150], "real": [152], "deployment": [153], "scenarios": [154], "rather": [155], "than": [156], "simulation.": [157], "We": [158], "compare": [159], "approaches": [162], "centralized": [165], "learning": [166], "baseline": [167], "across": [168], "performance": [170], "metrics": [171], "such": [172], "accuracy,": [175], "time,": [177], "communication": [178], "overhead,": [179], "energy": [181], "usage.": [182], "also": [185], "explores": [186], "various": [187], "workflows,": [189], "comparing": [190], "client-controlled": [191], "workflows": [192], "server-controlled": [194], "workflows.": [195], "findings": [197], "this": [199], "work": [200], "serve": [201], "valuable": [204], "reference": [205], "understanding": [207], "methods": [216], "segmentation": [218], "tasks": [219], "UAV-based": [221], "imaging.": [222]}, "primary_topic": {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.9994999766349792, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.9994999766349792, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11307", "display_name": "Domain Adaptation and Few-Shot Learning", "score": 9.999999747378752e-05, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2511.00055", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2511.00055", "pdf_url": "https://arxiv.org/pdf/2511.00055", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W7216499742", "doi": "https://doi.org/10.64758/ndsn8y87", "title": "A Hybrid Deep Learning Approach for Enhanced Intrusion Detection in Industrial Control Systems Using Federated Learning", "publication_year": 2025, "publication_date": "2025-01-01", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5140269258", "display_name": "Indu Sharma", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": true, "raw_author_name": "Dr. Indu Sharma", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Industrial": [0], "Control": [1], "Systems": [2, 13], "(ICS)": [3], "are": [4], "increasingly": [5], "vulnerable": [6], "to": [7, 52, 78, 168], "cyberattacks,": [8], "necessitating": [9], "robust": [10], "Intrusion": [11], "Detection": [12], "(IDS).": [14], "Traditional": [15], "IDS": [16], "approaches": [17], "often": [18], "struggle": [19], "with": [20, 110], "the": [21], "complexity": [22], "and": [23, 56, 162, 173, 196], "evolving": [24], "nature": [25], "of": [26, 137], "ICS": [27, 84, 131, 149, 190], "threats.": [28], "Deep": [29], "learning": [30, 66, 76, 156, 176], "(DL)": [31], "models": [32, 80, 172], "offer": [33], "promising": [34], "solutions,": [35], "but": [36], "their": [37], "performance": [38], "relies": [39], "heavily": [40], "on": [41, 126, 140, 146], "large,": [42], "centralized": [43, 170], "datasets,": [44], "which": [45], "may": [46], "be": [47], "impractical": [48], "or": [49], "infeasible": [50], "due": [51], "data": [53, 109, 180], "privacy": [54], "concerns": [55], "regulatory": [57], "constraints.": [58], "This": [59], "paper": [60], "proposes": [61], "a": [62, 92, 97, 111, 141, 147], "novel": [63], "hybrid": [64, 93, 154], "deep": [65], "approach": [67, 157], "for": [68, 102, 116], "enhanced": [69], "intrusion": [70], "detection": [71, 160, 195], "in": [72, 189], "ICS,": [73], "leveraging": [74], "federated": [75, 155], "(FL)": [77], "train": [79], "collaboratively": [81], "across": [82], "multiple": [83], "environments": [85], "without": [86], "sharing": [87], "sensitive": [88], "data.": [89], "We": [90], "develop": [91], "architecture": [94], "that": [95, 152], "combines": [96], "Convolutional": [98], "Neural": [99, 113], "Network": [100, 114], "(CNN)": [101], "feature": [103], "extraction": [104], "from": [105], "raw": [106], "network": [107], "traffic": [108], "Recurrent": [112], "(RNN)": [115], "capturing": [117], "temporal": [118], "dependencies.": [119], "The": [120, 182], "FL": [121], "framework": [122], "enables": [123], "distributed": [124], "training": [125], "local": [127], "datasets": [128], "within": [129], "each": [130], "site,": [132], "followed": [133], "by": [134], "secure": [135], "aggregation": [136], "model": [138], "updates": [139], "central": [142], "server.": [143], "Experimental": [144], "results": [145], "benchmark": [148], "dataset": [150], "demonstrate": [151], "our": [153], "achieves": [158], "superior": [159], "accuracy": [161], "lower": [163], "false": [164], "alarm": [165], "rates": [166], "compared": [167], "traditional": [169], "DL": [171], "conventional": [174], "machine": [175], "techniques,": [177], "while": [178], "preserving": [179], "privacy.": [181], "proposed": [183], "method": [184], "addresses": [185], "critical": [186], "security": [187], "challenges": [188], "environments,": [191], "enabling": [192], "proactive": [193], "threat": [194], "improved": [197], "overall": [198], "system": [199], "resilience.": [200]}, "primary_topic": {"id": "https://openalex.org/T10400", "display_name": "Network Security and Intrusion Detection", "score": 0.9961000084877014, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10400", "display_name": "Network Security and Intrusion Detection", "score": 0.9961000084877014, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.0010000000474974513, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10917", "display_name": "Smart Grid Security and Resilience", "score": 0.0003000000142492354, "subfield": {"id": "https://openalex.org/subfields/2207", "display_name": "Control and Systems Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:ojs2.janolijournals.com:article/9", "is_oa": false, "landing_page_url": "https://janolijournals.com/index.php/index/article/view/9", "pdf_url": null, "source": {"id": "https://openalex.org/S7407055700", "display_name": "JANOLI International Journal of Artificial Intelligence and its Applications", "issn_l": "3048-6815", "issn": ["3048-6815"], "is_oa": false, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": null, "host_organization_name": null, "host_organization_lineage": [], "host_organization_lineage_names": [], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "3048-6815", "raw_type": "info:eu-repo/semantics/article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4409153926", "doi": "https://doi.org/10.3390/designs9020045", "title": "A Novel Lightweight Deep Learning Approach for Drivers\u2019 Facial Expression Detection", "publication_year": 2025, "publication_date": "2025-04-03", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5036848523", "display_name": "Jia Uddin", "orcid": "https://orcid.org/0000-0002-3403-4095", "observed_orcids": ["https://orcid.org/0000-0002-3403-4095"]}, "institutions": [{"id": "https://openalex.org/I85389745", "display_name": "Woosong University", "ror": "https://ror.org/02srty072", "country_code": "KR", "type": "education", "lineage": ["https://openalex.org/I85389745"]}], "countries": ["KR"], "is_corresponding": true, "raw_author_name": "Jia Uddin", "raw_affiliation_strings": ["AI and Big Data Department, Endicott College, Woosong University, Daejeon 34606, Republic of Korea"], "raw_orcid": "https://orcid.org/0000-0002-3403-4095", "affiliations": [{"raw_affiliation_string": "AI and Big Data Department, Endicott College, Woosong University, Daejeon 34606, Republic of Korea", "institution_ids": ["https://openalex.org/I85389745"]}]}], "abstract_inverted_index": {"Drivers\u2019": [0], "facial": [1, 103], "expression": [2, 104], "recognition": [3, 234], "systems": [4, 30], "play": [5], "a": [6, 92, 114, 209, 215, 226], "pivotal": [7], "role": [8], "in": [9, 24, 31, 57, 212, 218, 235], "Advanced": [10], "Driver": [11], "Assistance": [12], "Systems": [13], "(ADASs)": [14], "by": [15, 168, 172, 196, 200], "monitoring": [16], "emotional": [17], "states": [18], "and": [19, 46, 128, 159, 170, 187, 198, 214, 231], "detecting": [20], "fatigue": [21], "or": [22], "distractions": [23], "real": [25], "time.": [26], "However,": [27], "deploying": [28], "such": [29], "resource-constrained": [32], "environments": [33], "like": [34], "vehicles": [35], "requires": [36], "lightweight": [37, 63], "architectures": [38, 64], "to": [39, 133], "ensure": [40], "real-time": [41], "performance,": [42], "efficient": [43, 230], "model": [44], "updates,": [45], "compatibility": [47], "with": [48, 83, 113, 163], "embedded": [49], "hardware.": [50], "Smaller": [51], "models": [52], "significantly": [53], "reduce": [54], "communication": [55], "overhead": [56], "distributed": [58], "training.": [59], "For": [60], "autonomous": [61], "vehicles,": [62], "also": [65], "minimize": [66], "the": [67, 110, 145, 175, 179], "data": [68], "transfer": [69], "required": [70], "for": [71, 78, 101, 229], "over-the-air": [72], "updates.": [73], "Moreover,": [74], "they": [75], "are": [76], "crucial": [77], "their": [79], "deployability": [80], "on": [81, 174, 202, 205], "hardware": [82], "limited": [84], "on-chip": [85], "memory.": [86], "In": [87], "this": [88], "work,": [89], "we": [90, 150], "propose": [91], "novel": [93], "Dual": [94, 115], "Attention": [95, 116, 126, 131], "Lightweight": [96], "Deep": [97], "Learning": [98], "(DALDL)": [99], "approach": [100, 108], "drivers\u2019": [102], "recognition.": [105], "The": [106], "proposed": [107], "combines": [109], "SqueezeNext": [111], "architecture": [112], "Convolution": [117], "(DAC)": [118], "block.": [119], "Our": [120], "DAC": [121], "block": [122], "integrates": [123], "Hybrid": [124], "Channel": [125], "(HCA)": [127], "Coordinate": [129], "Space": [130], "(CSA)": [132], "enhance": [134], "feature": [135], "extraction": [136], "efficiency": [137], "while": [138], "maintaining": [139], "minimal": [140], "parameter": [141], "overhead.": [142], "To": [143], "evaluate": [144], "effectiveness": [146], "of": [147], "our": [148], "architecture,": [149], "compare": [151], "it": [152, 182, 207], "against": [153], "two": [154], "baselines:": [155], "(a)": [156], "Vanilla": [157], "SqueezeNet": [158], "(b)": [160], "AlexNet.": [161], "Compared": [162], "SqueezeNet,": [164], "DALDL": [165, 193, 224], "improves": [166, 194], "accuracy": [167, 186, 195, 213], "7.96%": [169], "F1-score": [171, 199], "7.95%": [173], "KMU-FED": [176], "dataset.": [177], "On": [178], "CK+": [180], "dataset,": [181], "achieves": [183], "8.51%": [184], "higher": [185, 189], "8.40%": [188], "F1-score.": [190, 219], "Against": [191], "AlexNet,": [192], "4.34%": [197], "4.17%": [201], "KMU-FED.": [203], "Lastly,": [204], "CK+,": [206], "provides": [208], "5.36%": [210], "boost": [211], "7.24%": [216], "increase": [217], "These": [220], "results": [221], "demonstrate": [222], "that": [223], "is": [225], "promising": [227], "solution": [228], "accurate": [232], "emotion": [233], "real-world": [236], "automotive": [237], "applications.": [238]}, "primary_topic": {"id": "https://openalex.org/T10667", "display_name": "Emotion and Mood Recognition", "score": 0.9437999725341797, "subfield": {"id": "https://openalex.org/subfields/3205", "display_name": "Experimental and Cognitive Psychology"}, "field": {"id": "https://openalex.org/fields/32", "display_name": "Psychology"}, "domain": {"id": "https://openalex.org/domains/2", "display_name": "Social Sciences"}}, "topics": [{"id": "https://openalex.org/T10667", "display_name": "Emotion and Mood Recognition", "score": 0.9437999725341797, "subfield": {"id": "https://openalex.org/subfields/3205", "display_name": "Experimental and Cognitive Psychology"}, "field": {"id": "https://openalex.org/fields/32", "display_name": "Psychology"}, "domain": {"id": "https://openalex.org/domains/2", "display_name": "Social Sciences"}}, {"id": "https://openalex.org/T11448", "display_name": "Face recognition and analysis", "score": 0.029600000008940697, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.016699999570846558, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.3390/designs9020045", "is_oa": true, "landing_page_url": "https://doi.org/10.3390/designs9020045", "pdf_url": "https://www.mdpi.com/2411-9660/9/2/45/pdf?version=1743690041", "source": {"id": "https://openalex.org/S4210222353", "display_name": "Designs", "issn_l": "2411-9660", "issn": ["2411-9660"], "is_oa": true, "is_in_doaj": true, "is_core": true, "listed_in": ["cwts-core", "doaj", "norway-1"], "host_organization": "https://openalex.org/P4310310987", "host_organization_name": "Multidisciplinary Digital Publishing Institute", "host_organization_lineage": ["https://openalex.org/P4310310987"], "host_organization_lineage_names": ["Multidisciplinary Digital Publishing Institute"], "type": "journal"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Designs", "raw_type": "journal-article"}, "cited_by_count": 10, "is_retracted": false}, {"id": "https://openalex.org/W7108669370", "doi": "https://doi.org/10.5281/zenodo.17820535", "title": "Beyond Data and Model Parallelism: Exploiting Hierarchical Concurrency for Scalable Deep Learning", "publication_year": 2025, "publication_date": "2025-12-04", "type": "article", "language": null, "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Revista, Zen", "orcid": "https://orcid.org/0009-0007-6299-2008", "observed_orcids": ["https://orcid.org/0009-0007-6299-2008"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Revista, Zen", "raw_affiliation_strings": [], "raw_orcid": "https://orcid.org/0009-0007-6299-2008", "affiliations": []}, {"author_position": "last", "author": {"id": null, "display_name": "IA, 10", "orcid": "https://orcid.org/0009-0007-6299-2008", "observed_orcids": ["https://orcid.org/0009-0007-6299-2008"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "IA, 10", "raw_affiliation_strings": [], "raw_orcid": "https://orcid.org/0009-0007-6299-2008", "affiliations": []}], "abstract_inverted_index": {"Deep": [0], "learning": [1, 123, 153], "models": [2, 37, 154], "have": [3], "achieved": [4], "remarkable": [5], "success": [6], "in": [7, 27, 73, 128], "various": [8], "domains,": [9], "but": [10], "their": [11], "computational": [12, 87, 165], "demands": [13], "continue": [14], "to": [15, 50, 135, 146, 164], "grow": [16], "exponentially.": [17], "Traditional": [18], "data": [19, 137], "and": [20, 38, 70, 96, 113, 131, 138, 155], "model": [21, 139], "parallelism": [22, 72, 104, 140], "strategies": [23], "often": [24], "fall": [25], "short": [26], "addressing": [28], "the": [29, 52, 86, 101, 107, 110, 148], "scalability": [30], "challenges": [31], "posed": [32], "by": [33], "these": [34], "increasingly": [35], "complex": [36], "massive": [39], "datasets.": [40], "This": [41, 142], "paper": [42], "explores": [43], "hierarchical": [44, 75, 143], "concurrency": [45, 81], "as": [46], "a": [47, 64, 74], "novel": [48], "approach": [49, 144], "unlock": [51], "full": [53], "potential": [54], "of": [55, 80, 85, 103, 109, 150], "parallel": [56], "computing": [57], "resources": [58], "for": [59, 92], "deep": [60, 122, 152], "learning.": [61], "We": [62, 89], "propose": [63], "framework": [65], "that": [66, 98, 159], "combines": [67], "data,": [68, 112], "model,": [69, 111], "task": [71], "manner,": [76], "enabling": [77], "fine-grained": [78], "exploitation": [79], "at": [82], "different": [83], "levels": [84], "graph.": [88], "introduce": [90], "algorithms": [91], "adaptive": [93], "resource": [94, 132], "allocation": [95], "scheduling": [97], "dynamically": [99], "adjust": [100], "degree": [102], "based": [105], "on": [106, 120], "characteristics": [108], "hardware.": [114], "Furthermore,": [115], "we": [116], "present": [117], "experimental": [118], "results": [119], "large-scale": [121], "benchmarks,": [124], "demonstrating": [125], "significant": [126], "improvements": [127], "training": [129, 149], "time": [130], "utilization": [133], "compared": [134], "state-of-the-art": [136], "techniques.": [141], "promises": [145], "accelerate": [147], "future": [151], "enable": [156], "new": [157], "applications": [158], "were": [160], "previously": [161], "infeasible": [162], "due": [163], "limitations.": [166]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9794999957084656, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9794999957084656, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.006599999964237213, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.003800000064074993, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.5281/zenodo.17820535", "is_oa": true, "landing_page_url": "https://doi.org/10.5281/zenodo.17820535", "pdf_url": null, "source": {"id": "https://openalex.org/S4306400562", "display_name": "Zenodo (CERN European Organization for Nuclear Research)", "issn_l": null, "issn": null, "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I67311998", "host_organization_name": "European Organization for Nuclear Research", "host_organization_lineage": ["https://openalex.org/I67311998"], "host_organization_lineage_names": [], "type": "repository"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": null, "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "JournalArticle"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4413973325", "doi": "https://doi.org/10.14778/3746405.3746430", "title": "Effective and Efficient Distributed Temporal Graph Learning through Hotspot Memory Sharing", "publication_year": 2025, "publication_date": "2025-05-01", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5051865316", "display_name": "L.C. Zhang", "orcid": "https://orcid.org/0000-0001-8327-3175", "observed_orcids": ["https://orcid.org/0000-0001-8327-3175"]}, "institutions": [{"id": "https://openalex.org/I76130692", "display_name": "Zhejiang University", "ror": "https://ror.org/00a2xv884", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76130692"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Longjiao Zhang", "raw_affiliation_strings": ["Zhejiang University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Zhejiang University", "institution_ids": ["https://openalex.org/I76130692"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100431161", "display_name": "Rui Wang", "orcid": "https://orcid.org/0000-0001-8915-4169", "observed_orcids": ["https://orcid.org/0000-0001-8915-4169"]}, "institutions": [{"id": "https://openalex.org/I76130692", "display_name": "Zhejiang University", "ror": "https://ror.org/00a2xv884", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76130692"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Rui Wang", "raw_affiliation_strings": ["Zhejiang University, High-Tech Zone (Binjiang), Institute of Blockchain and Data Security"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Zhejiang University, High-Tech Zone (Binjiang), Institute of Blockchain and Data Security", "institution_ids": ["https://openalex.org/I76130692"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5058282223", "display_name": "Tongya Zheng", "orcid": "https://orcid.org/0000-0003-1190-9773", "observed_orcids": ["https://orcid.org/0000-0003-1190-9773"]}, "institutions": [{"id": "https://openalex.org/I4400573310", "display_name": "Hangzhou City University", "ror": "https://ror.org/01wck0s05", "country_code": null, "type": "education", "lineage": ["https://openalex.org/I4400573310"]}], "countries": [], "is_corresponding": false, "raw_author_name": "Tongya Zheng", "raw_affiliation_strings": ["Zhejiang Key Laboratory of Big Data Intelligent Computing, Hangzhou City University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Zhejiang Key Laboratory of Big Data Intelligent Computing, Hangzhou City University", "institution_ids": ["https://openalex.org/I4400573310"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5101830887", "display_name": "Ziqi Huang", "orcid": "https://orcid.org/0000-0001-8008-5873", "observed_orcids": ["https://orcid.org/0000-0001-8008-5873"]}, "institutions": [{"id": "https://openalex.org/I76130692", "display_name": "Zhejiang University", "ror": "https://ror.org/00a2xv884", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76130692"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Ziqi Huang", "raw_affiliation_strings": ["Zhejiang University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Zhejiang University", "institution_ids": ["https://openalex.org/I76130692"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5101638307", "display_name": "Wenjie Huang", "orcid": "https://orcid.org/0000-0003-3238-0573", "observed_orcids": ["https://orcid.org/0000-0003-3238-0573"]}, "institutions": [{"id": "https://openalex.org/I76130692", "display_name": "Zhejiang University", "ror": "https://ror.org/00a2xv884", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76130692"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Wenjie Huang", "raw_affiliation_strings": ["Zhejiang University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Zhejiang University", "institution_ids": ["https://openalex.org/I76130692"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100352800", "display_name": "Xinyu Wang", "orcid": "https://orcid.org/0000-0002-5507-6569", "observed_orcids": ["https://orcid.org/0000-0002-5507-6569"]}, "institutions": [{"id": "https://openalex.org/I76130692", "display_name": "Zhejiang University", "ror": "https://ror.org/00a2xv884", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76130692"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Xinyu Wang", "raw_affiliation_strings": ["Zhejiang University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Zhejiang University", "institution_ids": ["https://openalex.org/I76130692"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100428572", "display_name": "Can Wang", "orcid": "https://orcid.org/0000-0002-2890-0057", "observed_orcids": ["https://orcid.org/0000-0002-2890-0057", "https://orcid.org/0009-0004-6079-6829", "https://orcid.org/0009-0007-8596-5512"]}, "institutions": [{"id": "https://openalex.org/I76130692", "display_name": "Zhejiang University", "ror": "https://ror.org/00a2xv884", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76130692"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Can Wang", "raw_affiliation_strings": ["Zhejiang University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Zhejiang University", "institution_ids": ["https://openalex.org/I76130692"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5026532752", "display_name": "Mingli Song", "orcid": "https://orcid.org/0000-0003-2621-6048", "observed_orcids": ["https://orcid.org/0000-0003-2621-6048", "https://orcid.org/0009-0007-5092-0365"]}, "institutions": [{"id": "https://openalex.org/I76130692", "display_name": "Zhejiang University", "ror": "https://ror.org/00a2xv884", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76130692"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Mingli Song", "raw_affiliation_strings": ["Zhejiang University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Zhejiang University", "institution_ids": ["https://openalex.org/I76130692"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5017672363", "display_name": "Sai Wu", "orcid": "https://orcid.org/0000-0002-7903-1496", "observed_orcids": ["https://orcid.org/0000-0002-7903-1496", "https://orcid.org/0000-0002-1866-9197"]}, "institutions": [{"id": "https://openalex.org/I76130692", "display_name": "Zhejiang University", "ror": "https://ror.org/00a2xv884", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76130692"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Sai Wu", "raw_affiliation_strings": ["Zhejiang University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Zhejiang University", "institution_ids": ["https://openalex.org/I76130692"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5004997843", "display_name": "Shuibing He", "orcid": "https://orcid.org/0000-0002-7075-4153", "observed_orcids": ["https://orcid.org/0000-0002-7075-4153"]}, "institutions": [{"id": "https://openalex.org/I76130692", "display_name": "Zhejiang University", "ror": "https://ror.org/00a2xv884", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76130692"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Shuibing He", "raw_affiliation_strings": ["Zhejiang University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Zhejiang University", "institution_ids": ["https://openalex.org/I76130692"]}]}], "abstract_inverted_index": {"Memory-based": [0], "temporal": [1, 11, 22], "graph": [2, 110], "neural": [3], "network": [4], "(MTGNN)": [5], "models": [6], "are": [7], "effective": [8], "for": [9, 30, 101], "predicting": [10], "graphs": [12, 32], "by": [13], "using": [14], "node": [15, 82], "memory": [16, 47, 63, 83, 102], "and": [17, 23, 39, 46, 62, 95, 117, 134], "message-passing": [18], "modules": [19], "to": [20, 43, 97], "capture": [21], "structural": [24], "information,": [25], "respectively.": [26], "However,": [27], "distributed": [28, 74, 99, 129], "training": [29, 135], "large": [31], "presents": [33], "challenges": [34], "such": [35], "as": [36], "accuracy": [37, 133], "loss": [38], "decreased": [40], "efficiency": [41], "due": [42], "remote": [44], "features": [45], "transmission.": [48], "Despite": [49], "improvements": [50], "in": [51, 132], "MTGNN": [52, 75, 130], "system": [53], "optimizations,": [54], "issues": [55], "like": [56, 107], "dynamic": [57], "load": [58], "imbalances,": [59], "communication": [60, 100], "overhead,": [61], "staleness": [64], "persist.": [65], "To": [66], "tackle": [67], "these": [68], "challenges,": [69], "we": [70], "introduce": [71], "MemShare,": [72], "a": [73, 79, 87], "system.": [76], "MemShare": [77, 126], "introduces": [78], "novel": [80], "shared": [81, 91, 108, 112, 118], "paradigm": [84], "that": [85, 125], "utilizes": [86], "small": [88], "subset": [89], "of": [90], "nodes": [92], "across": [93], "machines": [94], "GPUs": [96], "reduce": [98], "management.": [103], "It": [104], "incorporates": [105], "techniques": [106], "nodes-centric": [109], "partitioning,": [111], "nodes-aware": [113], "boundary": [114], "decay": [115], "sampling,": [116], "nodes-targeted": [119], "synchronous": [120], "smoothing": [121], "aggregation.": [122], "Experiments": [123], "show": [124], "outperforms": [127], "existing": [128], "systems": [131], "efficiency.": [136]}, "primary_topic": {"id": "https://openalex.org/T11273", "display_name": "Advanced Graph Neural Networks", "score": 0.7372999787330627, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T11273", "display_name": "Advanced Graph Neural Networks", "score": 0.7372999787330627, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T12292", "display_name": "Graph Theory and Algorithms", "score": 0.23800000548362732, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11344", "display_name": "Traffic Prediction and Management Techniques", "score": 0.006200000178068876, "subfield": {"id": "https://openalex.org/subfields/2215", "display_name": "Building and Construction"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.14778/3746405.3746430", "is_oa": false, "landing_page_url": "https://doi.org/10.14778/3746405.3746430", "pdf_url": null, "source": {"id": "https://openalex.org/S4210226185", "display_name": "Proceedings of the VLDB Endowment", "issn_l": "2150-8097", "issn": ["2150-8097"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "norway-2"], "host_organization": "https://openalex.org/P4310319798", "host_organization_name": "Association for Computing Machinery", "host_organization_lineage": ["https://openalex.org/P4310319798"], "host_organization_lineage_names": ["Association for Computing Machinery"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Proceedings of the VLDB Endowment", "raw_type": "journal-article"}, "cited_by_count": 4, "is_retracted": false}, {"id": "https://openalex.org/W4417426823", "doi": null, "title": "Characterizing Communication Patterns in Distributed Large Language Model Inference", "publication_year": 2025, "publication_date": "2025-07-18", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5101847978", "display_name": "Lang Xu", "orcid": "https://orcid.org/0009-0000-6307-4566", "observed_orcids": ["https://orcid.org/0009-0000-6307-4566", "https://orcid.org/0009-0000-4984-494X"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Xu, Lang", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5055205226", "display_name": "Kaushik Kandadi Suresh", "orcid": "https://orcid.org/0000-0002-3705-2387", "observed_orcids": ["https://orcid.org/0000-0002-3705-2387"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Suresh, Kaushik Kandadi", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5015153122", "display_name": "Quentin G. Anthony", "orcid": "https://orcid.org/0000-0002-6823-9080", "observed_orcids": ["https://orcid.org/0000-0002-6823-9080"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Anthony, Quentin", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5061967476", "display_name": "Nawras Alnaasan", "orcid": "https://orcid.org/0000-0002-3638-4144", "observed_orcids": ["https://orcid.org/0000-0002-3638-4144"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Alnaasan, Nawras", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5024879682", "display_name": "Dhabaleswar K. DK Panda", "orcid": "https://orcid.org/0000-0002-0356-1781", "observed_orcids": ["https://orcid.org/0000-0002-0356-1781"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Panda, Dhabaleswar K.", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Large": [0], "Language": [1], "Models": [2], "(LLMs)": [3], "built": [4], "on": [5], "transformer": [6], "architectures": [7, 73], "have": [8], "transformed": [9], "natural": [10], "language": [11], "processing,": [12], "achieving": [13], "remarkable": [14], "performance": [15, 33], "across": [16, 93], "diverse": [17], "applications.": [18], "While": [19], "distributed": [20, 48], "inference": [21, 154], "frameworks": [22, 155], "enable": [23], "practical": [24, 137], "deployment": [25], "of": [26, 71], "these": [27], "models,": [28], "inter-GPU": [29], "communication": [30, 45, 91, 157], "creates": [31], "significant": [32], "constraints": [34], "that": [35, 99], "limit": [36], "service": [37], "quality": [38], "in": [39, 47, 76, 144], "real-world": [40], "systems.": [41], "This": [42], "paper": [43], "investigates": [44], "dynamics": [46], "LLM": [49, 146], "serving-analyzing": [50], "how": [51], "various": [52], "parallelization": [53, 95, 142], "approaches": [54, 126], "coordinate": [55], "data": [56, 117], "exchange": [57], "between": [58], "GPU": [59], "workers": [60], "during": [61], "inference.": [62], "We": [63], "study": [64], "dense": [65], "transformer-based": [66], "models": [67, 88], "as": [68], "representative": [69], "examples": [70], "contemporary": [72], "widely": [74], "used": [75], "operational": [77], "deployments.": [78], "Our": [79], "work": [80], "combines": [81], "detailed": [82], "profiling": [83], "measurements": [84], "with": [85], "predictive": [86], "analytical": [87], "to": [89, 130], "characterize": [90], "behavior": [92], "different": [94], "configurations.": [96], "Results": [97], "show": [98], "tensor": [100], "parallelism": [101, 115], "incurs": [102], "substantial": [103], "network": [104], "overhead": [105], "but": [106], "delivers": [107], "superior": [108], "response": [109], "times": [110], "for": [111, 139, 152], "brief": [112], "sequences,": [113], "pipeline": [114], "minimizes": [116], "transfer": [118], "requirements": [119], "while": [120], "increasing": [121], "total": [122], "latency,": [123], "and": [124, 148, 156], "combined": [125], "demand": [127], "careful": [128], "tuning": [129], "achieve": [131], "balanced": [132], "performance.": [133], "These": [134], "insights": [135], "offer": [136], "recommendations": [138], "selecting": [140], "appropriate": [141], "schemes": [143], "production": [145], "services": [147], "identify": [149], "key": [150], "opportunities": [151], "optimizing": [153], "infrastructure.": [158]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9434000253677368, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9434000253677368, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.012199999764561653, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10181", "display_name": "Natural Language Processing Techniques", "score": 0.009499999694526196, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2507.14392", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2507.14392", "pdf_url": "https://arxiv.org/pdf/2507.14392", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4415084608", "doi": "https://doi.org/10.48550/arxiv.2503.01328", "title": "PipeOffload: Improving Scalability of Pipeline Parallelism with Memory Optimization", "publication_year": 2025, "publication_date": "2025-03-03", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Wan, Xinyi", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Wan, Xinyi", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5078214310", "display_name": "Penghui Qi", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Qi, Penghui", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5034264575", "display_name": "Guangxing Huang", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Huang, Guangxing", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100616299", "display_name": "Min Lin", "orcid": "https://orcid.org/0000-0001-7042-9601", "observed_orcids": ["https://orcid.org/0000-0001-7042-9601"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Lin, Min", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5108050453", "display_name": "Jialin Li", "orcid": "https://orcid.org/0000-0003-4972-3480", "observed_orcids": ["https://orcid.org/0000-0003-4972-3480", "https://orcid.org/0009-0003-1111-5129", "https://orcid.org/0009-0007-6487-8997", "https://orcid.org/0000-0001-8076-7858"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Li, Jialin", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Pipeline": [0], "parallelism": [1], "(PP)": [2], "is": [3, 15, 86, 157], "widely": [4], "used": [5], "for": [6], "training": [7], "large": [8], "language": [9], "models": [10], "(LLMs),": [11], "yet": [12], "its": [13], "scalability": [14], "often": [16], "constrained": [17], "by": [18, 44], "high": [19], "activation": [20, 99, 127], "memory": [21, 48, 100, 108, 119, 128, 153], "consumption": [22], "as": [23], "the": [24, 31, 46, 60, 72, 81, 125, 132], "number": [25, 134], "of": [26, 33, 62, 71, 135], "in-flight": [27], "microbatches": [28], "grows": [29], "with": [30, 77, 110, 131, 150], "degree": [32], "PP.": [34, 52], "In": [35, 80], "this": [36, 42], "paper,": [37], "we": [38, 56, 89, 106], "focus": [39], "on": [40], "addressing": [41], "challenge": [43], "leveraging": [45], "under-explored": [47], "offload": [49, 94, 109], "strategy": [50, 95], "in": [51, 59, 101], "With": [53], "empirical": [54], "study,": [55], "discover": [57], "that": [58, 96, 124], "majority": [61], "standard": [63], "configurations,": [64], "at": [65, 159], "least": [66], "half,": [67], "and": [68, 118], "potentially": [69], "all,": [70], "activations": [73], "can": [74], "be": [75], "offloaded": [76], "negligible": [78], "overhead.": [79], "cases": [82], "where": [83], "full": [84], "overload": [85], "not": [87], "possible,": [88], "introduce": [90], "a": [91, 102, 139, 147], "novel": [92], "selective": [93], "decreases": [97], "peak": [98], "better-than-linear": [103], "manner.": [104], "Furthermore,": [105], "integrate": [107], "other": [111], "techniques": [112], "to": [113, 146], "jointly": [114], "consider": [115], "overall": [116], "throughput": [117], "limitation.": [120], "Our": [121], "experiments": [122], "proves": [123], "per-device": [126], "effectively": [129], "reduces": [130], "total": [133], "stages,": [136], "making": [137], "PP": [138], "stronger": [140], "alternative": [141], "than": [142], "TP,": [143], "offering": [144], "up": [145], "19\\%": [148], "acceleration": [149], "even": [151], "lower": [152], "consumption.": [154], "The": [155], "implementation": [156], "open-sourced": [158], "\\href{https://github.com/sail-sg/zero-bubble-pipeline-parallelism}{this": [160], "url}.": [161]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9589999914169312, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9589999914169312, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.006500000134110451, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.004600000102072954, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2503.01328", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2503.01328", "pdf_url": "https://arxiv.org/pdf/2503.01328", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4413967041", "doi": "https://doi.org/10.1109/tpds.2025.3605491", "title": "DynPipe: Toward Dynamic End-to-End Pipeline Parallelism for Interference-Aware DNN Training", "publication_year": 2025, "publication_date": "2025-09-03", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Zhengyi Yuan", "orcid": "https://orcid.org/0009-0000-6532-7176", "observed_orcids": ["https://orcid.org/0009-0000-6532-7176"]}, "institutions": [{"id": "https://openalex.org/I47720641", "display_name": "Huazhong University of Science and Technology", "ror": "https://ror.org/00p991c53", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I47720641"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Zhengyi Yuan", "raw_affiliation_strings": ["National Engineering Research Center for Big Data Technology and System, Services Computing Technology and System Lab/Cluster and Grid Computing Lab, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China"], "raw_orcid": "https://orcid.org/0009-0000-6532-7176", "affiliations": [{"raw_affiliation_string": "National Engineering Research Center for Big Data Technology and System, Services Computing Technology and System Lab/Cluster and Grid Computing Lab, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China", "institution_ids": ["https://openalex.org/I47720641"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5115596416", "display_name": "Xiong Wang", "orcid": "https://orcid.org/0000-0001-5360-0454", "observed_orcids": ["https://orcid.org/0000-0001-5360-0454"]}, "institutions": [{"id": "https://openalex.org/I47720641", "display_name": "Huazhong University of Science and Technology", "ror": "https://ror.org/00p991c53", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I47720641"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Xiong Wang", "raw_affiliation_strings": ["National Engineering Research Center for Big Data Technology and System, Services Computing Technology and System Lab/Cluster and Grid Computing Lab, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China"], "raw_orcid": "https://orcid.org/0000-0001-5360-0454", "affiliations": [{"raw_affiliation_string": "National Engineering Research Center for Big Data Technology and System, Services Computing Technology and System Lab/Cluster and Grid Computing Lab, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China", "institution_ids": ["https://openalex.org/I47720641"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Yuntao Nie", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I47720641", "display_name": "Huazhong University of Science and Technology", "ror": "https://ror.org/00p991c53", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I47720641"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yuntao Nie", "raw_affiliation_strings": ["National Engineering Research Center for Big Data Technology and System, Services Computing Technology and System Lab/Cluster and Grid Computing Lab, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "National Engineering Research Center for Big Data Technology and System, Services Computing Technology and System Lab/Cluster and Grid Computing Lab, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China", "institution_ids": ["https://openalex.org/I47720641"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5048609852", "display_name": "Yufei Tao", "orcid": "https://orcid.org/0000-0003-3883-5452", "observed_orcids": ["https://orcid.org/0000-0003-3883-5452"]}, "institutions": [{"id": "https://openalex.org/I47720641", "display_name": "Huazhong University of Science and Technology", "ror": "https://ror.org/00p991c53", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I47720641"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yufei Tao", "raw_affiliation_strings": ["National Engineering Research Center for Big Data Technology and System, Services Computing Technology and System Lab/Cluster and Grid Computing Lab, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "National Engineering Research Center for Big Data Technology and System, Services Computing Technology and System Lab/Cluster and Grid Computing Lab, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China", "institution_ids": ["https://openalex.org/I47720641"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100379940", "display_name": "Yuqing Li", "orcid": "https://orcid.org/0000-0003-0816-5777", "observed_orcids": ["https://orcid.org/0000-0003-0816-5777"]}, "institutions": [{"id": "https://openalex.org/I37461747", "display_name": "Wuhan University", "ror": "https://ror.org/033vjfk17", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I37461747"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yuqing Li", "raw_affiliation_strings": ["School of Cyber Science and Engineering, Wuhan University, Wuhan, China"], "raw_orcid": "https://orcid.org/0000-0003-0816-5777", "affiliations": [{"raw_affiliation_string": "School of Cyber Science and Engineering, Wuhan University, Wuhan, China", "institution_ids": ["https://openalex.org/I37461747"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5066761150", "display_name": "Zhiyuan Shao", "orcid": "https://orcid.org/0000-0003-2139-6465", "observed_orcids": ["https://orcid.org/0000-0003-2139-6465"]}, "institutions": [{"id": "https://openalex.org/I47720641", "display_name": "Huazhong University of Science and Technology", "ror": "https://ror.org/00p991c53", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I47720641"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Zhiyuan Shao", "raw_affiliation_strings": ["National Engineering Research Center for Big Data Technology and System, Services Computing Technology and System Lab/Cluster and Grid Computing Lab, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China"], "raw_orcid": "https://orcid.org/0000-0003-2139-6465", "affiliations": [{"raw_affiliation_string": "National Engineering Research Center for Big Data Technology and System, Services Computing Technology and System Lab/Cluster and Grid Computing Lab, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China", "institution_ids": ["https://openalex.org/I47720641"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5022398389", "display_name": "Xiaofei Liao", "orcid": "https://orcid.org/0000-0001-6302-813X", "observed_orcids": ["https://orcid.org/0000-0001-6302-813X"]}, "institutions": [{"id": "https://openalex.org/I47720641", "display_name": "Huazhong University of Science and Technology", "ror": "https://ror.org/00p991c53", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I47720641"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Xiaofei Liao", "raw_affiliation_strings": ["National Engineering Research Center for Big Data Technology and System, Services Computing Technology and System Lab/Cluster and Grid Computing Lab, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China"], "raw_orcid": "https://orcid.org/0000-0001-6302-813X", "affiliations": [{"raw_affiliation_string": "National Engineering Research Center for Big Data Technology and System, Services Computing Technology and System Lab/Cluster and Grid Computing Lab, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China", "institution_ids": ["https://openalex.org/I47720641"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100374506", "display_name": "Bo Li", "orcid": "https://orcid.org/0000-0003-2083-9105", "observed_orcids": ["https://orcid.org/0000-0003-2083-9105"]}, "institutions": [{"id": "https://openalex.org/I200769079", "display_name": "Hong Kong University of Science and Technology", "ror": "https://ror.org/00q4vv597", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I200769079"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Bo Li", "raw_affiliation_strings": ["Department of Computer Science and Engineering, Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong", "Department of Computer Science and Engineering, Hong Kong University of Science and Technology, Hong Kong"], "raw_orcid": "https://orcid.org/0000-0003-2083-9105", "affiliations": [{"raw_affiliation_string": "Department of Computer Science and Engineering, Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong", "institution_ids": ["https://openalex.org/I200769079"]}, {"raw_affiliation_string": "Department of Computer Science and Engineering, Hong Kong University of Science and Technology, Hong Kong", "institution_ids": ["https://openalex.org/I200769079"]}]}, {"author_position": "last", "author": {"id": null, "display_name": "Hai Jin", "orcid": "https://orcid.org/0000-0002-3934-7605", "observed_orcids": ["https://orcid.org/0000-0002-3934-7605"]}, "institutions": [{"id": "https://openalex.org/I47720641", "display_name": "Huazhong University of Science and Technology", "ror": "https://ror.org/00p991c53", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I47720641"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Hai Jin", "raw_affiliation_strings": ["National Engineering Research Center for Big Data Technology and System, Services Computing Technology and System Lab/Cluster and Grid Computing Lab, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China"], "raw_orcid": "https://orcid.org/0000-0002-3934-7605", "affiliations": [{"raw_affiliation_string": "National Engineering Research Center for Big Data Technology and System, Services Computing Technology and System Lab/Cluster and Grid Computing Lab, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China", "institution_ids": ["https://openalex.org/I47720641"]}]}], "abstract_inverted_index": {"Pipeline": [0], "parallelism": [1], "has": [2], "emerged": [3], "as": [4, 97], "an": [5, 66], "indispensable": [6], "technique": [7], "for": [8], "training": [9, 105], "large": [10], "deep": [11], "neural": [12], "networks.": [13], "While": [14], "existing": [15], "asynchronous": [16], "pipeline": [17, 45, 68, 125], "systems": [18], "address": [19], "the": [20, 72, 91, 104, 108, 139], "time": [21], "bubbles": [22], "inherent": [23], "in": [24, 51, 127], "synchronous": [25], "architectures,": [26], "they": [27], "continue": [28], "to": [29, 34, 47, 89, 114], "suffer": [30], "frominefficiencyandsusceptibilitytovolatilehardware": [31], "environment": [32], "due": [33], "their": [35], "suboptimal": [36], "andstaticconfigurations.": [37], "In": [38], "this": [39], "paper,": [40], "we": [41], "propose": [42], "DynPipe,": [43], "aninterference-awareasynchronous": [44], "framework": [46], "optimize": [48], "theend-to-endtraining": [49], "performance": [50], "highlydynamiccomputing": [52], "environments.": [53, 129], "By": [54], "characterizing": [55], "thenon-overlappedcommunication": [56], "overheads": [57], "andconvergencerate": [58], "conditioned": [59], "on": [60, 103], "stage-wise": [61], "staleness,": [62], "DynPipe": [63, 79, 111, 134], "carefully": [64], "crafts": [65], "optimized": [67], "partition": [69], "that": [70, 84, 133], "harmonizes": [71], "hardware": [73], "speed": [74], "with": [75], "statistical": [76], "convergence.": [77], "Moreover,": [78], "deploys": [80], "anon-intrusiverandom": [81], "forest": [82], "model": [83], "utilizes": [85], "runtime": [86], "stage": [87], "statistics": [88], "evaluate": [90], "impact": [92], "of": [93], "environmental": [94], "changes,": [95], "such": [96], "task": [98], "interference": [99], "and": [100, 118], "network": [101], "jitter,": [102], "efficiency.": [106], "Following": [107], "evaluation": [109], "guidance,": [110], "adaptivelyadjustspartition": [112], "plan": [113], "restore": [115], "both": [116], "intra": [117], "inter-stage": [119], "load": [120], "balancing,": [121], "thereby": [122], "facilitating": [123], "seamless": [124], "reconfiguration": [126], "dynamic": [128], "Extensive": [130], "experiments": [131], "show": [132], "outperforms": [135], "state-of-the-art": [136], "systems,": [137], "accelerating": [138], "time-to-accuracy": [140], "by1.5-3.4\u00d7.": [141]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9761000275611877, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9761000275611877, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.004000000189989805, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.00279999990016222, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/tpds.2025.3605491", "is_oa": true, "landing_page_url": "https://doi.org/10.1109/tpds.2025.3605491", "pdf_url": null, "source": {"id": "https://openalex.org/S97130795", "display_name": "IEEE Transactions on Parallel and Distributed Systems", "issn_l": "1045-9219", "issn": ["1045-9219", "1558-2183", "2161-9883"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-2", "norway-2"], "host_organization": "https://openalex.org/P4310319808", "host_organization_name": "Institute of Electrical and Electronics Engineers", "host_organization_lineage": ["https://openalex.org/P4310319808"], "host_organization_lineage_names": ["Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Transactions on Parallel and Distributed Systems", "raw_type": "journal-article"}, "cited_by_count": 5, "is_retracted": false}, {"id": "https://openalex.org/W4410952128", "doi": "https://doi.org/10.3390/app15116251", "title": "Low-Scalability Distributed Systems for Artificial Intelligence: A Comparative Study of Distributed Deep Learning Frameworks for Image Classification", "publication_year": 2025, "publication_date": "2025-06-02", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5117799499", "display_name": "Manuel Rivera-Escobedo", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I47400469", "display_name": "Universidad Aut\u00f3noma de Zacatecas \"Francisco Garc\u00eda Salinas\"", "ror": "https://ror.org/01m296r74", "country_code": "MX", "type": "education", "lineage": ["https://openalex.org/I47400469"]}], "countries": ["MX"], "is_corresponding": false, "raw_author_name": "Manuel Rivera-Escobedo", "raw_affiliation_strings": ["Laboratorio de Invenciones Aplicadas a la Industria (LIAI), Unidad Acad\u00e9mica de Ingenier\u00eda El\u00e9ctrica, Universidad Aut\u00f3noma de Zacatecas, Zacatecas 98000, Mexico"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Laboratorio de Invenciones Aplicadas a la Industria (LIAI), Unidad Acad\u00e9mica de Ingenier\u00eda El\u00e9ctrica, Universidad Aut\u00f3noma de Zacatecas, Zacatecas 98000, Mexico", "institution_ids": ["https://openalex.org/I47400469"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5103091150", "display_name": "Manuel de Jes\u00fas L\u00f3pez-Mart\u00ednez", "orcid": "https://orcid.org/0000-0003-3063-7718", "observed_orcids": ["https://orcid.org/0000-0003-3063-7718"]}, "institutions": [{"id": "https://openalex.org/I47400469", "display_name": "Universidad Aut\u00f3noma de Zacatecas \"Francisco Garc\u00eda Salinas\"", "ror": "https://ror.org/01m296r74", "country_code": "MX", "type": "education", "lineage": ["https://openalex.org/I47400469"]}], "countries": ["MX"], "is_corresponding": true, "raw_author_name": "Manuel de Jes\u00fas L\u00f3pez-Mart\u00ednez", "raw_affiliation_strings": ["Laboratorio de Invenciones Aplicadas a la Industria (LIAI), Unidad Acad\u00e9mica de Ingenier\u00eda El\u00e9ctrica, Universidad Aut\u00f3noma de Zacatecas, Zacatecas 98000, Mexico"], "raw_orcid": "https://orcid.org/0000-0003-3063-7718", "affiliations": [{"raw_affiliation_string": "Laboratorio de Invenciones Aplicadas a la Industria (LIAI), Unidad Acad\u00e9mica de Ingenier\u00eda El\u00e9ctrica, Universidad Aut\u00f3noma de Zacatecas, Zacatecas 98000, Mexico", "institution_ids": ["https://openalex.org/I47400469"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5061981327", "display_name": "Luis Octavio Sol\u00eds-S\u00e1nchez", "orcid": "https://orcid.org/0000-0003-2545-4116", "observed_orcids": ["https://orcid.org/0000-0003-2545-4116"]}, "institutions": [{"id": "https://openalex.org/I47400469", "display_name": "Universidad Aut\u00f3noma de Zacatecas \"Francisco Garc\u00eda Salinas\"", "ror": "https://ror.org/01m296r74", "country_code": "MX", "type": "education", "lineage": ["https://openalex.org/I47400469"]}], "countries": ["MX"], "is_corresponding": false, "raw_author_name": "Luis Octavio Solis-S\u00e1nchez", "raw_affiliation_strings": ["Laboratorio de Sistemas Inteligentes de Visi\u00f3n Artificial, Posgrado en Ingenier\u00eda y Tecnolog\u00eda Aplicada, Universidad Aut\u00f3noma de Zacatecas, Zacatecas 98000, Mexico"], "raw_orcid": "https://orcid.org/0000-0003-2545-4116", "affiliations": [{"raw_affiliation_string": "Laboratorio de Sistemas Inteligentes de Visi\u00f3n Artificial, Posgrado en Ingenier\u00eda y Tecnolog\u00eda Aplicada, Universidad Aut\u00f3noma de Zacatecas, Zacatecas 98000, Mexico", "institution_ids": ["https://openalex.org/I47400469"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5077686955", "display_name": "H\u00e9ctor A. Guerrero-Osuna", "orcid": "https://orcid.org/0000-0003-0743-842X", "observed_orcids": ["https://orcid.org/0000-0003-0743-842X"]}, "institutions": [{"id": "https://openalex.org/I47400469", "display_name": "Universidad Aut\u00f3noma de Zacatecas \"Francisco Garc\u00eda Salinas\"", "ror": "https://ror.org/01m296r74", "country_code": "MX", "type": "education", "lineage": ["https://openalex.org/I47400469"]}], "countries": ["MX"], "is_corresponding": false, "raw_author_name": "H\u00e9ctor Alonso Guerrero-Osuna", "raw_affiliation_strings": ["Unidad Acad\u00e9mica de Ingenier\u00eda El\u00e9ctrica, Universidad Aut\u00f3noma de Zacatecas, Zacatecas 98000, Mexico"], "raw_orcid": "https://orcid.org/0000-0003-0743-842X", "affiliations": [{"raw_affiliation_string": "Unidad Acad\u00e9mica de Ingenier\u00eda El\u00e9ctrica, Universidad Aut\u00f3noma de Zacatecas, Zacatecas 98000, Mexico", "institution_ids": ["https://openalex.org/I47400469"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5022598506", "display_name": "Sodel V\u00e1zquez-Reyes", "orcid": "https://orcid.org/0000-0002-9435-5499", "observed_orcids": ["https://orcid.org/0000-0002-9435-5499"]}, "institutions": [{"id": "https://openalex.org/I47400469", "display_name": "Universidad Aut\u00f3noma de Zacatecas \"Francisco Garc\u00eda Salinas\"", "ror": "https://ror.org/01m296r74", "country_code": "MX", "type": "education", "lineage": ["https://openalex.org/I47400469"]}], "countries": ["MX"], "is_corresponding": false, "raw_author_name": "Sodel V\u00e1zquez-Reyes", "raw_affiliation_strings": ["Unidad Acad\u00e9mica de Ingenier\u00eda El\u00e9ctrica, Universidad Aut\u00f3noma de Zacatecas, Zacatecas 98000, Mexico"], "raw_orcid": "https://orcid.org/0000-0002-9435-5499", "affiliations": [{"raw_affiliation_string": "Unidad Acad\u00e9mica de Ingenier\u00eda El\u00e9ctrica, Universidad Aut\u00f3noma de Zacatecas, Zacatecas 98000, Mexico", "institution_ids": ["https://openalex.org/I47400469"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5117799500", "display_name": "Daniel Acosta-Escare\u00f1o", "orcid": "https://orcid.org/0000-0002-9271-8727", "observed_orcids": ["https://orcid.org/0000-0002-9271-8727"]}, "institutions": [{"id": "https://openalex.org/I47400469", "display_name": "Universidad Aut\u00f3noma de Zacatecas \"Francisco Garc\u00eda Salinas\"", "ror": "https://ror.org/01m296r74", "country_code": "MX", "type": "education", "lineage": ["https://openalex.org/I47400469"]}], "countries": ["MX"], "is_corresponding": false, "raw_author_name": "Daniel Acosta-Escare\u00f1o", "raw_affiliation_strings": ["Unidad Acad\u00e9mica de Ingenier\u00eda El\u00e9ctrica, Universidad Aut\u00f3noma de Zacatecas, Zacatecas 98000, Mexico"], "raw_orcid": "https://orcid.org/0000-0002-9271-8727", "affiliations": [{"raw_affiliation_string": "Unidad Acad\u00e9mica de Ingenier\u00eda El\u00e9ctrica, Universidad Aut\u00f3noma de Zacatecas, Zacatecas 98000, Mexico", "institution_ids": ["https://openalex.org/I47400469"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5001784673", "display_name": "Carlos Alberto Olvera-Olvera", "orcid": "https://orcid.org/0000-0003-1519-7718", "observed_orcids": ["https://orcid.org/0000-0003-1519-7718"]}, "institutions": [{"id": "https://openalex.org/I47400469", "display_name": "Universidad Aut\u00f3noma de Zacatecas \"Francisco Garc\u00eda Salinas\"", "ror": "https://ror.org/01m296r74", "country_code": "MX", "type": "education", "lineage": ["https://openalex.org/I47400469"]}], "countries": ["MX"], "is_corresponding": true, "raw_author_name": "Carlos A. Olvera-Olvera", "raw_affiliation_strings": ["Laboratorio de Invenciones Aplicadas a la Industria (LIAI), Unidad Acad\u00e9mica de Ingenier\u00eda El\u00e9ctrica, Universidad Aut\u00f3noma de Zacatecas, Zacatecas 98000, Mexico"], "raw_orcid": "https://orcid.org/0000-0003-1519-7718", "affiliations": [{"raw_affiliation_string": "Laboratorio de Invenciones Aplicadas a la Industria (LIAI), Unidad Acad\u00e9mica de Ingenier\u00eda El\u00e9ctrica, Universidad Aut\u00f3noma de Zacatecas, Zacatecas 98000, Mexico", "institution_ids": ["https://openalex.org/I47400469"]}]}], "abstract_inverted_index": {"Artificial": [0], "intelligence": [1, 33, 94, 189], "has": [2, 17], "experienced": [3], "tremendous": [4], "growth": [5], "in": [6, 12, 104, 207, 225], "various": [7], "areas": [8, 227], "of": [9, 27, 46, 83, 101, 124, 134], "knowledge,": [10], "especially": [11], "computer": [13], "science.": [14], "Distributed": [15, 139, 143], "computing": [16, 103, 224], "become": [18], "necessary": [19], "for": [20, 30, 53, 130, 156], "storing,": [21], "processing,": [22], "and": [23, 35, 68, 107, 142, 147, 159, 171, 183, 198, 204, 232], "generating": [24], "large": [25, 44], "amounts": [26, 45], "information": [28], "essential": [29], "training": [31, 133], "artificial": [32, 93, 188], "models": [34], "algorithms": [36], "that": [37, 78, 178], "allow": [38], "knowledge": [39], "to": [40, 80, 90, 212, 219, 222], "be": [41, 193], "created": [42], "from": [43], "data.": [47], "Currently,": [48], "cloud": [49], "services": [50, 74], "offer": [51], "products": [52], "running": [54], "distributed": [55, 102, 113, 132, 185], "data": [56], "training,": [57], "such": [58, 167, 228], "as": [59, 153, 168, 229], "NVIDIA": [60], "Deep": [61], "Learning": [62], "Solutions,": [63], "Amazon": [64], "SageMaker,": [65], "Microsoft": [66], "Azure,": [67], "Google": [69, 169], "Cloud": [70], "AI": [71], "Platform.": [72], "These": [73], "have": [75], "a": [76, 111, 154, 172], "cost": [77], "adapts": [79], "the": [81, 99, 125, 131, 150, 201], "needs": [82], "users": [84], "who": [85], "require": [86], "high": [87], "processing": [88, 106], "performance": [89, 161], "perform": [91], "their": [92, 160], "tasks.": [95], "This": [96], "study": [97], "highlights": [98], "relevance": [100], "image": [105, 157], "classification": [108], "tasks": [109, 190], "using": [110, 149], "low-scalability": [112], "system": [114], "built": [115], "with": [116, 164, 180], "devices": [117], "considered": [118], "obsolete.": [119], "To": [120], "this": [121, 208], "end,": [122], "two": [123], "most": [126], "widely": [127], "used": [128], "libraries": [129], "deep": [135], "learning": [136], "models,": [137], "PyTorch\u2019s": [138], "Data": [140], "Parallel": [141], "TensorFlow,": [144], "were": [145], "implemented": [146], "evaluated": [148], "ResNet50": [151], "model": [152], "basis": [155], "classification,": [158], "was": [162], "compared": [163], "modern": [165], "environments": [166], "Colab": [170], "recent": [173], "Workstation.": [174], "The": [175], "results": [176, 202], "demonstrate": [177], "even": [179], "low": [181], "scalability": [182], "outdated": [184], "systems,": [186], "comprehensive": [187], "can": [191], "still": [192], "performed,": [194], "reducing": [195], "investment": [196], "time": [197], "costs.": [199], "With": [200], "obtained": [203], "experiments": [205], "conducted": [206], "study,": [209], "we": [210], "aim": [211], "promote": [213], "technological": [214], "sustainability": [215], "through": [216], "device": [217], "recycling": [218], "facilitate": [220], "access": [221], "high-performance": [223], "key": [226], "research,": [230], "industry,": [231], "education.": [233]}, "primary_topic": {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.7964000105857849, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.7964000105857849, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.0812000036239624, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.032099999487400055, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.3390/app15116251", "is_oa": true, "landing_page_url": "https://doi.org/10.3390/app15116251", "pdf_url": "https://www.mdpi.com/2076-3417/15/11/6251/pdf?version=1748851945", "source": {"id": "https://openalex.org/S4210205812", "display_name": "Applied Sciences", "issn_l": "2076-3417", "issn": ["2076-3417"], "is_oa": true, "is_in_doaj": true, "is_core": true, "listed_in": ["cwts-core", "doaj", "norway-1"], "host_organization": "https://openalex.org/P4310310987", "host_organization_name": "Multidisciplinary Digital Publishing Institute", "host_organization_lineage": ["https://openalex.org/P4310310987"], "host_organization_lineage_names": ["Multidisciplinary Digital Publishing Institute"], "type": "journal"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Applied Sciences", "raw_type": "journal-article"}, "cited_by_count": 3, "is_retracted": false}, {"id": "https://openalex.org/W4406140140", "doi": "https://doi.org/10.36227/techrxiv.172469825.57827560/v3", "title": "Revolutionizing Optical Networks: The Integration and Impact of Large Language Models", "publication_year": 2025, "publication_date": "2025-01-07", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5093793521", "display_name": "Sergio Cruzes", "orcid": "https://orcid.org/0009-0008-8955-7628", "observed_orcids": ["https://orcid.org/0009-0008-8955-7628"]}, "institutions": [], "countries": [], "is_corresponding": true, "raw_author_name": "Sergio Cruzes", "raw_affiliation_strings": [], "raw_orcid": "https://orcid.org/0009-0008-8955-7628", "affiliations": []}], "abstract_inverted_index": {"Large": [0], "Language": [1], "Models": [2], "(LLMs),": [3], "built": [4], "on": [5], "the": [6, 16, 122, 285, 348], "transformer": [7], "architecture,": [8], "are": [9, 64, 132], "revolutionizing": [10], "optical": [11, 35, 142, 291, 344], "network": [12, 82, 152, 297, 345], "management": [13], "by": [14, 24], "addressing": [15], "complexities": [17], "of": [18, 70, 97, 288], "these": [19, 202], "highly": [20], "specialized": [21], "systems.": [22, 358], "Characterized": [23], "real-time": [25, 151, 193], "performance": [26], "requirements,": [27], "multi-vendor": [28], "equipment": [29], "interoperability,": [30], "and": [31, 51, 59, 75, 88, 90, 105, 113, 121, 134, 156, 170, 198, 217, 225, 234, 255, 301, 316, 326, 355], "intricate": [32], "signal": [33], "processing,": [34], "networks": [36, 143], "face": [37, 179], "challenges": [38, 55, 180, 203], "such": [39, 181, 206], "as": [40, 182, 207], "managing": [41], "diverse": [42, 331], "transmission": [43], "impairments,": [44], "including": [45, 311], "chromatic": [46], "dispersion,": [47, 50], "polarization": [48], "mode": [49], "nonlinear": [52], "effects.": [53], "These": [54], "demand": [56], "advanced": [57, 107], "automation": [58], "optimization": [60, 158], "techniques,": [61], "which": [62], "LLMs": [63, 71, 110, 117, 145, 178, 289, 320, 338], "well-suited": [65], "to": [66, 78, 124, 222, 342], "address.": [67], "The": [68, 138, 304], "integration": [69, 312], "provides": [72], "a": [73, 334, 340], "scalable": [74], "adaptive": [76, 235, 356], "approach": [77], "automating": [79], "tasks": [80, 194, 332], "like": [81, 195, 229, 263], "configuration,": [83], "fault": [84, 199], "diagnosis,": [85], "alarm": [86, 299], "management,": [87], "routing": [89], "spectral": [91], "assignment": [92], "(RSA).": [93], "By": [94, 329], "enhancing": [95], "Quality": [96], "Transmission": [98], "(QoT)": [99], "estimation,": [100], "optimizing": [101, 242], "amplifier": [102], "gain": [103], "control,": [104], "supporting": [106], "simulation": [108], "frameworks,": [109], "enable": [111], "efficient": [112], "dynamic": [114, 196], "decision-making.": [115, 174], "Additionally,": [116], "offer": [118, 339], "user-friendly": [119], "interfaces": [120], "ability": [123], "incorporate": [125], "Human-in-the-Loop": [126], "(HITL)": [127], "systems,": [128], "ensuring": [129], "critical": [130], "decisions": [131], "monitored": [133], "managed": [135], "in": [136, 192, 290, 296], "real-time.": [137], "proposed": [139], "framework": [140], "for": [141, 350], "combines": [144], "with": [146, 214, 308, 313], "digital": [147, 314], "twin": [148], "technology,": [149], "enabling": [150], "monitoring,": [153], "predictive": [154], "analysis,": [155], "scenariobased": [157], "within": [159], "virtualized": [160], "environments.": [161], "This": [162, 282], "synergy": [163], "reduces": [164], "operational": [165], "complexity,": [166], "enhances": [167], "resource": [168, 243, 302], "efficiency,": [169], "facilitates": [171], "intelligent,": [172, 353], "autonomous": [173], "Despite": [175], "their": [176, 294], "promise,": [177], "hallucinations\u2014producing": [183], "semantically": [184], "incorrect": [185], "or": [186, 270], "fabricated": [187], "outputs\u2014and": [188], "computational": [189], "latency,": [190], "particularly": [191], "reconfiguration": [197], "resolution.": [200], "Addressing": [201], "requires": [204], "strategies": [205, 262], "prompt": [208], "engineering,": [209], "retrieval-augmented": [210], "generation": [211], "(RAG),": [212], "fine-tuning": [213], "domainspecific": [215], "data,": [216], "integrating": [218], "error": [219], "prediction": [220], "models": [221], "improve": [223], "accuracy": [224], "reduce": [226, 276], "risks.": [227], "Techniques": [228], "edge": [230], "computing,": [231], "model": [232, 272], "pruning,": [233], "deployment": [236, 318], "help": [237], "mitigate": [238], "latency": [239], "concerns": [240], "while": [241], "utilization.": [244], "Energy": [245], "efficiency": [246], "is": [247], "further": [248], "enhanced": [249], "through": [250], "hardware": [251], "optimization,": [252], "algorithm": [253], "refinement,": [254], "renewable": [256], "energy": [257, 277, 327], "adoption.": [258], "Studies": [259], "demonstrate": [260, 306], "that": [261, 307], "lowering": [264], "graphics": [265], "grocessing": [266], "units": [267], "(GPU)": [268], "frequencies": [269], "employing": [271], "parallelism": [273], "can": [274, 321], "significantly": [275, 322], "consumption": [278], "without": [279], "sacrificing": [280], "performance.": [281], "paper": [283], "explores": [284], "transformative": [286], "potential": [287], "networks,": [292], "highlighting": [293], "applications": [295], "design,": [298], "compression,": [300], "optimization.": [303], "results": [305], "proper": [309], "adaptation,": [310], "twins": [315], "sustainable": [317], "strategies,": [319], "enhance": [323], "automation,": [324], "scalability,": [325], "efficiency.": [328], "unifying": [330], "under": [333], "single": [335], "intelligent": [336], "framework,": [337], "pathway": [341], "revolutionize": [343], "architectures,": [346], "paving": [347], "way": [349], "greener,": [351], "more": [352], "resilient,": [354], "communication": [357]}, "primary_topic": {"id": "https://openalex.org/T10847", "display_name": "Advanced Optical Network Technologies", "score": 0.6054999828338623, "subfield": {"id": "https://openalex.org/subfields/2208", "display_name": "Electrical and Electronic Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10847", "display_name": "Advanced Optical Network Technologies", "score": 0.6054999828338623, "subfield": {"id": "https://openalex.org/subfields/2208", "display_name": "Electrical and Electronic Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10232", "display_name": "Optical Network Technologies", "score": 0.2653999924659729, "subfield": {"id": "https://openalex.org/subfields/2208", "display_name": "Electrical and Electronic Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10714", "display_name": "Software-Defined Networks and 5G", "score": 0.05209999904036522, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.36227/techrxiv.172469825.57827560/v3", "is_oa": true, "landing_page_url": "https://doi.org/10.36227/techrxiv.172469825.57827560/v3", "pdf_url": null, "source": null, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "acceptedVersion", "is_accepted": true, "is_published": false, "raw_source_name": null, "raw_type": "posted-content"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W7108711514", "doi": "https://doi.org/10.5281/zenodo.17819420", "title": "Scalable AI Through Hybrid Parallelism: Balancing Data and Model Distribution for Enhanced Performance", "publication_year": 2025, "publication_date": "2025-12-04", "type": "article", "language": null, "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Revista, Zen", "orcid": "https://orcid.org/0009-0007-6299-2008", "observed_orcids": ["https://orcid.org/0009-0007-6299-2008"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Revista, Zen", "raw_affiliation_strings": [], "raw_orcid": "https://orcid.org/0009-0007-6299-2008", "affiliations": []}, {"author_position": "last", "author": {"id": null, "display_name": "IA, 10", "orcid": "https://orcid.org/0009-0007-6299-2008", "observed_orcids": ["https://orcid.org/0009-0007-6299-2008"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "IA, 10", "raw_affiliation_strings": [], "raw_orcid": "https://orcid.org/0009-0007-6299-2008", "affiliations": []}], "abstract_inverted_index": {"This": [0, 62], "paper": [1, 108], "investigates": [2], "the": [3, 34, 96, 107, 116, 125, 128, 131, 134, 137, 153, 169, 191, 196], "challenges": [4], "and": [5, 20, 32, 39, 50, 78, 92, 103, 120, 136, 155, 171, 175, 183, 195], "opportunities": [6], "in": [7, 30], "scaling": [8], "Artificial": [9], "Intelligence": [10], "(AI)": [11], "through": [12], "hybrid": [13, 69, 147], "parallelism,": [14, 45, 52, 70], "specifically": [15], "focusing": [16], "on": [17, 124], "balancing": [18], "data": [19, 48, 73, 119, 163], "model": [21, 51, 76, 121, 165], "distribution": [22], "for": [23, 37, 67, 113, 190], "enhanced": [24], "performance.": [25], "As": [26], "AI": [27, 90, 129, 158, 178, 194, 200], "models": [28, 91, 159, 179], "grow": [29], "size": [31, 132], "complexity,": [33], "computational": [35, 99, 139], "demands": [36], "training": [38, 170], "inference": [40], "become": [41], "increasingly": [42], "prohibitive.": [43], "Hybrid": [44], "which": [46], "combines": [47], "parallelism": [49, 122, 148, 166], "offers": [53], "a": [54, 110], "promising": [55], "approach": [56], "to": [57, 161], "address": [58], "these": [59, 84, 188], "scalability": [60, 154], "challenges.": [61], "study": [63], "explores": [64], "various": [65], "strategies": [66, 85], "implementing": [68], "including": [71], "optimal": [72], "partitioning,": [74], "efficient": [75], "sharding,": [77], "communication-reducing": [79], "techniques.": [80], "The": [81, 185], "performance": [82, 156], "of": [83, 118, 127, 133, 157, 173, 187, 193, 198], "is": [86], "evaluated": [87], "across": [88], "different": [89], "hardware": [93], "architectures,": [94], "highlighting": [95], "trade-offs": [97], "between": [98], "efficiency,": [100], "memory": [101], "usage,": [102], "communication": [104], "overhead.": [105], "Furthermore,": [106], "presents": [109], "novel": [111], "framework": [112, 149], "dynamically": [114], "adapting": [115], "degree": [117], "based": [123], "characteristics": [126], "model,": [130], "dataset,": [135], "available": [138], "resources.": [140], "Experimental": [141], "results": [142], "demonstrate": [143], "that": [144], "this": [145], "adaptive": [146], "can": [150], "significantly": [151], "improve": [152], "compared": [160], "traditional": [162], "or": [164], "alone,": [167], "enabling": [168], "deployment": [172], "larger": [174], "more": [176], "complex": [177], "with": [180], "reduced": [181], "time": [182], "cost.": [184], "implications": [186], "findings": [189], "future": [192], "development": [197], "scalable": [199], "systems": [201], "are": [202], "also": [203], "discussed.": [204]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9854000210762024, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9854000210762024, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.0035000001080334187, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.002099999925121665, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.5281/zenodo.17819420", "is_oa": true, "landing_page_url": "https://doi.org/10.5281/zenodo.17819420", "pdf_url": null, "source": {"id": "https://openalex.org/S4306400562", "display_name": "Zenodo (CERN European Organization for Nuclear Research)", "issn_l": null, "issn": null, "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I67311998", "host_organization_name": "European Organization for Nuclear Research", "host_organization_lineage": ["https://openalex.org/I67311998"], "host_organization_lineage_names": [], "type": "repository"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": null, "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "JournalArticle"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4408182811", "doi": "https://doi.org/10.1109/tcc.2025.3548604", "title": "Deadline-Aware Online Job Scheduling for Distributed Training in Heterogeneous Clusters", "publication_year": 2025, "publication_date": "2025-03-06", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5100424312", "display_name": "Yuchen Zhang", "orcid": "https://orcid.org/0000-0002-3153-4000", "observed_orcids": ["https://orcid.org/0000-0002-3153-4000"]}, "institutions": [{"id": "https://openalex.org/I150229711", "display_name": "University of Electronic Science and Technology of China", "ror": "https://ror.org/04qr3zq92", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I150229711"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yuchen Zhang", "raw_affiliation_strings": ["University of Electronic Science and Technology of China, Chengdu, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "University of Electronic Science and Technology of China, Chengdu, China", "institution_ids": ["https://openalex.org/I150229711"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5015538526", "display_name": "Long Luo", "orcid": "https://orcid.org/0000-0003-1028-8178", "observed_orcids": ["https://orcid.org/0000-0003-1028-8178"]}, "institutions": [{"id": "https://openalex.org/I150229711", "display_name": "University of Electronic Science and Technology of China", "ror": "https://ror.org/04qr3zq92", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I150229711"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Long Luo", "raw_affiliation_strings": ["University of Electronic Science and Technology of China, Chengdu, China"], "raw_orcid": "https://orcid.org/0000-0003-1028-8178", "affiliations": [{"raw_affiliation_string": "University of Electronic Science and Technology of China, Chengdu, China", "institution_ids": ["https://openalex.org/I150229711"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5053945147", "display_name": "Gang Sun", "orcid": "https://orcid.org/0000-0002-2448-8915", "observed_orcids": ["https://orcid.org/0000-0002-2448-8915", "https://orcid.org/0009-0000-1622-5288"]}, "institutions": [{"id": "https://openalex.org/I150229711", "display_name": "University of Electronic Science and Technology of China", "ror": "https://ror.org/04qr3zq92", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I150229711"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Gang Sun", "raw_affiliation_strings": ["University of Electronic Science and Technology of China, Chengdu, China"], "raw_orcid": "https://orcid.org/0000-0002-2448-8915", "affiliations": [{"raw_affiliation_string": "University of Electronic Science and Technology of China, Chengdu, China", "institution_ids": ["https://openalex.org/I150229711"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5101507232", "display_name": "Hongfang Yu", "orcid": "https://orcid.org/0000-0002-5219-1780", "observed_orcids": ["https://orcid.org/0000-0002-5219-1780"]}, "institutions": [{"id": "https://openalex.org/I150229711", "display_name": "University of Electronic Science and Technology of China", "ror": "https://ror.org/04qr3zq92", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I150229711"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Hongfang Yu", "raw_affiliation_strings": ["University of Electronic Science and Technology of China, Chengdu, China"], "raw_orcid": "https://orcid.org/0000-0002-5219-1780", "affiliations": [{"raw_affiliation_string": "University of Electronic Science and Technology of China, Chengdu, China", "institution_ids": ["https://openalex.org/I150229711"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5100374506", "display_name": "Bo Li", "orcid": "https://orcid.org/0000-0003-2083-9105", "observed_orcids": ["https://orcid.org/0000-0003-2083-9105"]}, "institutions": [{"id": "https://openalex.org/I200769079", "display_name": "Hong Kong University of Science and Technology", "ror": "https://ror.org/00q4vv597", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I200769079"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Bo Li", "raw_affiliation_strings": ["Hong Kong University of Science and Technology, Hong Kong", "Hong Kong University of Science and Technology, China"], "raw_orcid": "https://orcid.org/0000-0003-2083-9105", "affiliations": [{"raw_affiliation_string": "Hong Kong University of Science and Technology, Hong Kong", "institution_ids": ["https://openalex.org/I200769079"]}, {"raw_affiliation_string": "Hong Kong University of Science and Technology, China", "institution_ids": ["https://openalex.org/I200769079"]}]}], "abstract_inverted_index": {"The": [0], "explosive": [1], "growth": [2], "in": [3, 18, 27, 51, 62, 111, 147], "training": [4, 25, 100], "data": [5], "and": [6, 122, 161], "model": [7], "sizes": [8], "has": [9], "spurred": [10], "the": [11, 87, 90, 104, 124, 130, 179], "adoption": [12], "of": [13, 92, 106, 132], "distributed": [14, 24], "deep": [15], "learning": [16], "(DL)": [17], "heterogeneous": [19, 29, 63, 112], "computing": [20], "clusters.": [21, 64, 114], "Efficiently": [22], "scheduling": [23, 125], "jobs": [26, 107, 134], "such": [28], "environments": [30], "while": [31], "ensuring": [32], "they": [33, 54], "meet": [34], "user-specified": [35], "deadlines": [36, 61, 110], "remains": [37], "a": [38, 79, 155], "critical": [39], "challenge.": [40], "While": [41], "most": [42], "existing": [43], "works": [44], "focus": [45], "on": [46, 169], "reducing": [47], "job": [48, 60, 97, 117], "completion": [49], "time": [50], "homogeneous": [52], "clusters,": [53], "pay": [55], "little": [56], "attention": [57], "to": [58, 95, 102, 184], "meeting": [59, 108], "To": [65, 142], "address": [66], "this": [67, 144], "issue,": [68], "we": [69, 149], "proposeDancer(Deadline-Aware": [70], "dyNamiC": [71], "GPU": [72, 113], "allocation": [73, 121], "approach": [74], "for": [75, 128], "Efficient": [76], "Resource": [77], "utilization),": [78], "novel": [80], "framework": [81], "that": [82], "dynamically": [83], "adjusts": [84], "not": [85], "only": [86], "number": [88, 105, 131], "but": [89], "type": [91], "GPUs": [93], "assigned": [94], "each": [96], "throughout": [98], "its": [99], "lifecycle.Danceraims": [101], "maximize": [103], "their": [109], "It": [115], "decouples": [116], "placement": [118], "from": [119], "resource": [120], "formulates": [123], "optimization": [126], "problem": [127, 146], "maximizing": [129], "deadline-meeting": [133], "as": [135], "an": [136, 151], "Integer": [137], "Linear": [138], "Programming": [139], "(ILP)": [140], "problem.": [141], "solve": [143], "ILP": [145], "real-time,": [148], "propose": [150], "online": [152], "algorithm": [153], "with": [154], "competitive": [156], "ratio": [157, 182], "guarantee,": [158], "leveraging": [159], "primal-dual": [160], "dynamic": [162], "programming": [163], "techniques.": [164], "Extensive": [165], "trace-driven": [166], "simulations": [167], "based": [168], "real-world": [170], "DL": [171], "workloads": [172], "demonstrate": [173], "thatDancersignificantly": [174], "outperforms": [175], "state-of-the-art": [176], "approaches,": [177], "improving": [178], "deadline": [180], "satisfactory": [181], "up": [183], "58.9%\u201374.2%.": [185]}, "primary_topic": {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.8080999851226807, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.8080999851226807, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10715", "display_name": "Distributed and Parallel Computing Systems", "score": 0.13269999623298645, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.014600000344216824, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/tcc.2025.3548604", "is_oa": false, "landing_page_url": "https://doi.org/10.1109/tcc.2025.3548604", "pdf_url": null, "source": {"id": "https://openalex.org/S2492498579", "display_name": "IEEE Transactions on Cloud Computing", "issn_l": "2168-7161", "issn": ["2168-7161", "2372-0018"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-1", "norway-1"], "host_organization": "https://openalex.org/P4310319808", "host_organization_name": "Institute of Electrical and Electronics Engineers", "host_organization_lineage": ["https://openalex.org/P4310319808"], "host_organization_lineage_names": ["Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Transactions on Cloud Computing", "raw_type": "journal-article"}, "cited_by_count": 13, "is_retracted": false}, {"id": "https://openalex.org/W4414644745", "doi": "https://doi.org/10.48550/arxiv.2504.16068", "title": "High-performance training and inference for deep equivariant interatomic potentials", "publication_year": 2025, "publication_date": "2025-04-22", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5042392407", "display_name": "Chuin Wei Tan", "orcid": "https://orcid.org/0000-0002-0087-7537", "observed_orcids": ["https://orcid.org/0000-0002-0087-7537"]}, "institutions": [{"id": "https://openalex.org/I136199984", "display_name": "Harvard University", "ror": "https://ror.org/03vek6s52", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I136199984"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Tan, Chuin Wei", "raw_affiliation_strings": ["John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA", "institution_ids": ["https://openalex.org/I136199984"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5031379596", "display_name": "Marc Descoteaux", "orcid": "https://orcid.org/0000-0001-8504-3429", "observed_orcids": ["https://orcid.org/0000-0001-8504-3429"]}, "institutions": [{"id": "https://openalex.org/I136199984", "display_name": "Harvard University", "ror": "https://ror.org/03vek6s52", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I136199984"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Descoteaux, Marc L.", "raw_affiliation_strings": ["John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA", "institution_ids": ["https://openalex.org/I136199984"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5119774372", "display_name": "Mit Kotak", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I63966007", "display_name": "Massachusetts Institute of Technology", "ror": "https://ror.org/042nb2s44", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I63966007"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Kotak, Mit", "raw_affiliation_strings": ["Center for Computational Science and Engineering , Massachusetts Institute of Technology , Cambridge , MA , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Center for Computational Science and Engineering , Massachusetts Institute of Technology , Cambridge , MA , USA", "institution_ids": ["https://openalex.org/I63966007"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Nascimento, Gabriel de Miranda", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Nascimento, Gabriel de Miranda", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5054195094", "display_name": "Se\u00e1n R. Kavanagh", "orcid": "https://orcid.org/0000-0003-4577-9647", "observed_orcids": ["https://orcid.org/0000-0003-4577-9647"]}, "institutions": [{"id": "https://openalex.org/I136199984", "display_name": "Harvard University", "ror": "https://ror.org/03vek6s52", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I136199984"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Kavanagh, Se\u00e1n R.", "raw_affiliation_strings": ["Center for the Environment , Harvard University , Cambridge , MA , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Center for the Environment , Harvard University , Cambridge , MA , USA", "institution_ids": ["https://openalex.org/I136199984"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5008670548", "display_name": "Laura Zichi", "orcid": "https://orcid.org/0000-0003-3897-3097", "observed_orcids": ["https://orcid.org/0000-0003-3897-3097"]}, "institutions": [{"id": "https://openalex.org/I136199984", "display_name": "Harvard University", "ror": "https://ror.org/03vek6s52", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I136199984"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Zichi, Laura", "raw_affiliation_strings": ["John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA", "institution_ids": ["https://openalex.org/I136199984"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5051222422", "display_name": "M. J. Wang", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I136199984", "display_name": "Harvard University", "ror": "https://ror.org/03vek6s52", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I136199984"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Wang, Menghang", "raw_affiliation_strings": ["John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA", "institution_ids": ["https://openalex.org/I136199984"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5119774373", "display_name": "Aadit Saluja", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I136199984", "display_name": "Harvard University", "ror": "https://ror.org/03vek6s52", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I136199984"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Saluja, Aadit", "raw_affiliation_strings": ["John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA", "institution_ids": ["https://openalex.org/I136199984"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Hu, Yizhong R.", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I136199984", "display_name": "Harvard University", "ror": "https://ror.org/03vek6s52", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I136199984"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Hu, Yizhong R.", "raw_affiliation_strings": ["John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA", "institution_ids": ["https://openalex.org/I136199984"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5062301881", "display_name": "Tess Smidt", "orcid": "https://orcid.org/0000-0001-5581-5344", "observed_orcids": ["https://orcid.org/0000-0001-5581-5344"]}, "institutions": [{"id": "https://openalex.org/I63966007", "display_name": "Massachusetts Institute of Technology", "ror": "https://ror.org/042nb2s44", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I63966007"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Smidt, Tess", "raw_affiliation_strings": ["Research Laboratory of Electronics , Massachusetts Institute of Technology , Cambridge , MA , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Research Laboratory of Electronics , Massachusetts Institute of Technology , Cambridge , MA , USA", "institution_ids": ["https://openalex.org/I63966007"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Johansson, Anders", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I4210104735", "display_name": "Sandia National Laboratories", "ror": "https://ror.org/01apwpt12", "country_code": "US", "type": "facility", "lineage": ["https://openalex.org/I1330989302", "https://openalex.org/I198811213", "https://openalex.org/I4210104735"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Johansson, Anders", "raw_affiliation_strings": ["Sandia National Laboratories , Albuquerque , NM , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Sandia National Laboratories , Albuquerque , NM , USA", "institution_ids": ["https://openalex.org/I4210104735"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5087328377", "display_name": "William C. Witt", "orcid": "https://orcid.org/0000-0002-1578-1888", "observed_orcids": ["https://orcid.org/0000-0002-1578-1888"]}, "institutions": [{"id": "https://openalex.org/I136199984", "display_name": "Harvard University", "ror": "https://ror.org/03vek6s52", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I136199984"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Witt, William C.", "raw_affiliation_strings": ["John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA", "institution_ids": ["https://openalex.org/I136199984"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5075010416", "display_name": "Boris Kozinsky", "orcid": "https://orcid.org/0000-0002-0638-539X", "observed_orcids": ["https://orcid.org/0000-0002-0638-539X"]}, "institutions": [{"id": "https://openalex.org/I136199984", "display_name": "Harvard University", "ror": "https://ror.org/03vek6s52", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I136199984"]}, {"id": "https://openalex.org/I4210120115", "display_name": "Robert Bosch (United States)", "ror": "https://ror.org/02venad53", "country_code": "US", "type": "company", "lineage": ["https://openalex.org/I4210120115", "https://openalex.org/I889804353"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Kozinsky, Boris", "raw_affiliation_strings": ["John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA", "Robert Bosch LLC Research and Technology Center , Watertown , MA , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA", "institution_ids": ["https://openalex.org/I136199984"]}, {"raw_affiliation_string": "Robert Bosch LLC Research and Technology Center , Watertown , MA , USA", "institution_ids": ["https://openalex.org/I4210120115"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5035916730", "display_name": "Albert Musaelian", "orcid": "https://orcid.org/0000-0003-0475-9236", "observed_orcids": ["https://orcid.org/0000-0003-0475-9236"]}, "institutions": [{"id": "https://openalex.org/I136199984", "display_name": "Harvard University", "ror": "https://ror.org/03vek6s52", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I136199984"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Musaelian, Albert", "raw_affiliation_strings": ["John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA", "Mirian Technologies Inc. , Boston , MA , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "John A. Paulson School of Engineering and Applied Sciences , Harvard University , Cambridge , MA , USA", "institution_ids": ["https://openalex.org/I136199984"]}, {"raw_affiliation_string": "Mirian Technologies Inc. , Boston , MA , USA", "institution_ids": []}]}], "abstract_inverted_index": {"Machine": [0], "learning": [1, 126], "interatomic": [2, 127], "potentials,": [3], "particularly": [4], "those": [5], "based": [6], "on": [7, 58, 71, 100, 153], "deep": [8], "equivariant": [9], "neural": [10], "networks,": [11], "have": [12], "demonstrated": [13], "state-of-the-art": [14], "accuracy": [15], "and": [16, 26, 33, 43, 63, 74], "computational": [17, 61], "efficiency": [18], "in": [19, 92], "atomistic": [20], "modeling": [21], "tasks": [22], "like": [23], "molecular": [24, 107, 150], "dynamics": [25, 151], "high-throughput": [27], "screening.": [28], "The": [29, 65], "size": [30], "of": [31, 35, 53, 80, 105, 156, 164], "datasets": [32, 73], "demands": [34], "downstream": [36], "workflows": [37], "are": [38], "growing": [39], "rapidly,": [40], "making": [41], "robust": [42], "scalable": [44], "software": [45], "essential.": [46], "This": [47], "work": [48], "presents": [49], "a": [50, 93, 132, 162], "major": [51], "overhaul": [52], "the": [54, 81, 101, 113, 119, 136, 142], "NequIP": [55], "framework": [56, 67], "focusing": [57], "multi-node": [59], "parallelism,": [60], "performance,": [62], "extensibility.": [64], "redesigned": [66], "supports": [68], "distributed": [69], "training": [70, 97], "large": [72], "removes": [75], "barriers": [76], "preventing": [77], "full": [78], "utilization": [79], "PyTorch": [82, 120], "2.0": [83], "compiler": [84, 123], "at": [85], "train": [86], "time.": [87], "We": [88], "demonstrate": [89], "this": [90], "acceleration": [91], "case": [94], "study": [95], "by": [96, 159], "Allegro": [98, 137], "models": [99], "SPICE": [102], "2": [103], "dataset": [104], "organic": [106], "systems.": [108], "For": [109], "inference,": [110], "we": [111, 130], "introduce": [112], "first": [114], "end-to-end": [115], "infrastructure": [116], "that": [117], "uses": [118], "Ahead-of-Time": [121], "Inductor": [122], "for": [124, 135], "machine": [125], "potentials.": [128], "Additionally,": [129], "implement": [131], "custom": [133], "kernel": [134], "model's": [138], "most": [139], "expensive": [140], "operation,": [141], "tensor": [143], "product.": [144], "Together,": [145], "these": [146], "advancements": [147], "speed": [148], "up": [149, 160], "calculations": [152], "system": [154], "sizes": [155], "practical": [157], "relevance": [158], "to": [161], "factor": [163], "18.": [165]}, "primary_topic": {"id": "https://openalex.org/T11948", "display_name": "Machine Learning in Materials Science", "score": 0.9854000210762024, "subfield": {"id": "https://openalex.org/subfields/2505", "display_name": "Materials Chemistry"}, "field": {"id": "https://openalex.org/fields/25", "display_name": "Materials Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T11948", "display_name": "Machine Learning in Materials Science", "score": 0.9854000210762024, "subfield": {"id": "https://openalex.org/subfields/2505", "display_name": "Materials Chemistry"}, "field": {"id": "https://openalex.org/fields/25", "display_name": "Materials Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10002", "display_name": "Advanced Chemical Physics Studies", "score": 0.008999999612569809, "subfield": {"id": "https://openalex.org/subfields/3107", "display_name": "Atomic and Molecular Physics, and Optics"}, "field": {"id": "https://openalex.org/fields/31", "display_name": "Physics and Astronomy"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.0012000000569969416, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2504.16068", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2504.16068", "pdf_url": "https://arxiv.org/pdf/2504.16068", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 11, "is_retracted": false}, {"id": "https://openalex.org/W4414586621", "doi": "https://doi.org/10.48550/arxiv.2505.19548", "title": "How Syntax Specialization Emerges in Language Models", "publication_year": 2025, "publication_date": "2025-05-26", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5045290666", "display_name": "Xufeng Duan", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I177725633", "display_name": "Chinese University of Hong Kong", "ror": "https://ror.org/00t33hh48", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I177725633"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Duan, Xufeng", "raw_affiliation_strings": ["Department of Linguistics and Modern Languages , The Chinese University of Hong Kong"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Linguistics and Modern Languages , The Chinese University of Hong Kong", "institution_ids": ["https://openalex.org/I177725633"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Yao, Zhaoqian", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I177725633", "display_name": "Chinese University of Hong Kong", "ror": "https://ror.org/00t33hh48", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I177725633"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Yao, Zhaoqian", "raw_affiliation_strings": ["Department of Linguistics and Modern Languages , The Chinese University of Hong Kong"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Linguistics and Modern Languages , The Chinese University of Hong Kong", "institution_ids": ["https://openalex.org/I177725633"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5031000779", "display_name": "Yunhao Zhang", "orcid": "https://orcid.org/0009-0009-3182-9296", "observed_orcids": ["https://orcid.org/0009-0009-3182-9296", "https://orcid.org/0009-0004-2029-7929", "https://orcid.org/0009-0009-0574-4395"]}, "institutions": [{"id": "https://openalex.org/I19820366", "display_name": "Chinese Academy of Sciences", "ror": "https://ror.org/034t30j35", "country_code": "CN", "type": "government", "lineage": ["https://openalex.org/I19820366"]}, {"id": "https://openalex.org/I4210112150", "display_name": "Institute of Automation", "ror": "https://ror.org/022c3hy66", "country_code": "CN", "type": "facility", "lineage": ["https://openalex.org/I19820366", "https://openalex.org/I4210112150"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Zhang, Yunhao", "raw_affiliation_strings": ["Institute of Automation , Chinese Academy of Sciences"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Institute of Automation , Chinese Academy of Sciences", "institution_ids": ["https://openalex.org/I19820366", "https://openalex.org/I4210112150"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5062722477", "display_name": "Shaonan Wang", "orcid": "https://orcid.org/0000-0001-5455-1359", "observed_orcids": ["https://orcid.org/0000-0001-5455-1359"]}, "institutions": [{"id": "https://openalex.org/I19820366", "display_name": "Chinese Academy of Sciences", "ror": "https://ror.org/034t30j35", "country_code": "CN", "type": "government", "lineage": ["https://openalex.org/I19820366"]}, {"id": "https://openalex.org/I4210112150", "display_name": "Institute of Automation", "ror": "https://ror.org/022c3hy66", "country_code": "CN", "type": "facility", "lineage": ["https://openalex.org/I19820366", "https://openalex.org/I4210112150"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Wang, Shaonan", "raw_affiliation_strings": ["Institute of Automation , Chinese Academy of Sciences"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Institute of Automation , Chinese Academy of Sciences", "institution_ids": ["https://openalex.org/I19820366", "https://openalex.org/I4210112150"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5077498783", "display_name": "Zhenguang Garry Cai", "orcid": "https://orcid.org/0000-0002-4097-6038", "observed_orcids": ["https://orcid.org/0000-0002-4097-6038"]}, "institutions": [{"id": "https://openalex.org/I177725633", "display_name": "Chinese University of Hong Kong", "ror": "https://ror.org/00t33hh48", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I177725633"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Cai, Zhenguang G.", "raw_affiliation_strings": ["Brain and Mind Institute , The Chinese University of Hong Kong", "Department of Linguistics and Modern Languages , The Chinese University of Hong Kong"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Brain and Mind Institute , The Chinese University of Hong Kong", "institution_ids": ["https://openalex.org/I177725633"]}, {"raw_affiliation_string": "Department of Linguistics and Modern Languages , The Chinese University of Hong Kong", "institution_ids": ["https://openalex.org/I177725633"]}]}], "abstract_inverted_index": {"Large": [0], "language": [1], "models": [2, 136], "(LLMs)": [3], "have": [4], "been": [5], "found": [6], "to": [7, 21], "develop": [8], "surprising": [9], "internal": [10, 68, 99], "specializations:": [11], "Individual": [12], "neurons,": [13], "attention": [14], "heads,": [15], "and": [16, 41, 92, 107, 113, 119, 151], "circuits": [17], "become": [18], "selectively": [19], "sensitive": [20], "syntactic": [22, 69, 76], "structure,": [23], "reflecting": [24], "patterns": [25], "observed": [26], "in": [27, 89, 130], "the": [28, 55, 148], "human": [29], "brain.": [30], "While": [31], "this": [32, 50], "specialization": [33, 59], "is": [34, 103, 114], "well-documented,": [35], "how": [36, 134], "it": [37, 138], "emerges": [38, 86], "during": [39, 139], "training": [40, 120, 152], "what": [42], "influences": [43], "its": [44, 62], "development": [45], "remains": [46], "largely": [47], "unknown.": [48], "In": [49], "work,": [51], "we": [52, 78, 145], "tap": [53], "into": [54], "black": [56], "box": [57], "of": [58, 97], "by": [60, 116], "tracking": [61], "formation": [63], "over": [64], "time.": [65], "By": [66], "quantifying": [67], "consistency": [70], "across": [71, 105], "minimal": [72], "pairs": [73], "from": [74], "various": [75], "phenomena,": [77], "identify": [79], "a": [80, 94], "clear": [81], "developmental": [82], "trajectory:": [83], "Syntactic": [84], "sensitivity": [85], "gradually,": [87], "concentrates": [88], "specific": [90], "layers,": [91], "exhibits": [93], "'critical": [95], "period'": [96], "rapid": [98], "specialization.": [100], "This": [101], "process": [102], "consistent": [104], "architectures": [106], "initialization": [108], "parameters": [109], "(e.g.,": [110], "random": [111], "seeds),": [112], "influenced": [115], "model": [117], "scale": [118], "data.": [121], "We": [122], "therefore": [123], "reveal": [124], "not": [125], "only": [126], "where": [127], "syntax": [128], "arises": [129], "LLMs": [131], "but": [132], "also": [133], "some": [135], "internalize": [137], "training.": [140], "To": [141], "support": [142], "future": [143], "research,": [144], "will": [146], "release": [147], "code,": [149], "models,": [150], "checkpoints": [153], "upon": [154], "acceptance.": [155]}, "primary_topic": {"id": "https://openalex.org/T10181", "display_name": "Natural Language Processing Techniques", "score": 0.4790000021457672, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10181", "display_name": "Natural Language Processing Techniques", "score": 0.4790000021457672, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.27219998836517334, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10465", "display_name": "Neurobiology of Language and Bilingualism", "score": 0.08479999750852585, "subfield": {"id": "https://openalex.org/subfields/2805", "display_name": "Cognitive Neuroscience"}, "field": {"id": "https://openalex.org/fields/28", "display_name": "Neuroscience"}, "domain": {"id": "https://openalex.org/domains/1", "display_name": "Life Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2505.19548", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2505.19548", "pdf_url": "https://arxiv.org/pdf/2505.19548", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": "cc-by-nc-sa", "license_id": "https://openalex.org/licenses/cc-by-nc-sa", "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4410454663", "doi": "https://doi.org/10.20944/preprints202505.1156.v1", "title": "Multimodal and Distributed LLMs: Bridging Scalability and Cross-Modal Reasoning", "publication_year": 2025, "publication_date": "2025-05-15", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Rajesh Kumar", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I35440088", "display_name": "ETH Zurich", "ror": "https://ror.org/05a28rw58", "country_code": "CH", "type": "education", "lineage": ["https://openalex.org/I2799323385", "https://openalex.org/I35440088"]}], "countries": ["CH"], "is_corresponding": false, "raw_author_name": "Rajesh Kumar", "raw_affiliation_strings": ["Department of Computer Science , ETH Zurich , Switzerland"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Computer Science , ETH Zurich , Switzerland", "institution_ids": ["https://openalex.org/I35440088"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Isabelle Laurent", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I39804081", "display_name": "Sorbonne Universit\u00e9", "ror": "https://ror.org/02en5vm52", "country_code": "FR", "type": "education", "lineage": ["https://openalex.org/I39804081"]}], "countries": ["FR"], "is_corresponding": false, "raw_author_name": "Isabelle Laurent", "raw_affiliation_strings": ["Department of Artificial Intelligence , Sorbonne University , France"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Artificial Intelligence , Sorbonne University , France", "institution_ids": ["https://openalex.org/I39804081"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "David M\u00fcller", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I100066346", "display_name": "University of Stuttgart", "ror": "https://ror.org/04vnq7t77", "country_code": "DE", "type": "education", "lineage": ["https://openalex.org/I100066346"]}], "countries": ["DE"], "is_corresponding": false, "raw_author_name": "David M\u00fcller", "raw_affiliation_strings": ["Department of Computer Science , University of Stuttgart , Germany"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Computer Science , University of Stuttgart , Germany", "institution_ids": ["https://openalex.org/I100066346"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5116857926", "display_name": "Klaus Elli", "orcid": "https://orcid.org/0009-0002-5673-6116", "observed_orcids": ["https://orcid.org/0009-0002-5673-6116"]}, "institutions": [{"id": "https://openalex.org/I100066346", "display_name": "University of Stuttgart", "ror": "https://ror.org/04vnq7t77", "country_code": "DE", "type": "education", "lineage": ["https://openalex.org/I100066346"]}, {"id": "https://openalex.org/I35440088", "display_name": "ETH Zurich", "ror": "https://ror.org/05a28rw58", "country_code": "CH", "type": "education", "lineage": ["https://openalex.org/I2799323385", "https://openalex.org/I35440088"]}], "countries": ["CH", "DE"], "is_corresponding": true, "raw_author_name": "Klaus Elli", "raw_affiliation_strings": ["Department of Computer Science , ETH Zurich , Switzerland", "Department of Computer Science , University of Stuttgart , Germany"], "raw_orcid": "https://orcid.org/0009-0002-5673-6116", "affiliations": [{"raw_affiliation_string": "Department of Computer Science , ETH Zurich , Switzerland", "institution_ids": ["https://openalex.org/I35440088"]}, {"raw_affiliation_string": "Department of Computer Science , University of Stuttgart , Germany", "institution_ids": ["https://openalex.org/I100066346"]}]}], "abstract_inverted_index": {"Large": [0], "Language": [1], "Models": [2], "(LLMs)": [3], "have": [4, 27, 34], "emerged": [5], "as": [6, 61, 118, 164, 291], "a": [7, 76, 258, 292, 296], "cornerstone": [8], "of": [9, 47, 58, 79, 84, 108, 205, 246, 264, 278, 305], "modern": [10], "artificial": [11], "intelligence,": [12], "achieving": [13], "remarkable": [14], "capabilities": [15], "in": [16, 82, 94, 202], "natural": [17], "language": [18, 70], "understanding": [19], "and": [20, 25, 31, 42, 54, 64, 92, 105, 110, 121, 140, 148, 160, 179, 185, 193, 212, 230, 236, 243, 272, 281, 295, 300, 308], "generation.": [21], "As": [22], "their": [23, 36], "scale": [24], "utility": [26], "increased,": [28], "two": [29], "critical": [30], "complementary": [32], "trends": [33], "defined": [35], "evolution:": [37], "(1)": [38], "the": [39, 56, 80, 90, 106, 187, 203, 244, 261, 303], "distributed": [40, 95, 192, 309], "systems": [41], "algorithms": [43], "enabling": [44], "efficient": [45], "training": [46, 129, 141], "ultra-large": [48], "models": [49, 71, 175], "across": [50, 150], "massive": [51], "compute": [52], "infrastructures,": [53], "(2)": [55], "integration": [57], "multiple": [59], "modalities\u2014such": [60], "vision,": [62], "audio,": [63], "structured": [65], "data\u2014into": [66], "unified": [67, 226], "multimodal": [68, 194, 241, 273], "large": [69], "(MLLMs).": [72], "This": [73, 287], "survey": [74, 288], "provides": [75], "comprehensive": [77], "examination": [78], "state-of-the-art": [81], "both": [83, 290], "these": [85], "dimensions.We": [86], "begin": [87], "by": [88], "exploring": [89], "foundations": [91], "advances": [93], "training,": [96], "including": [97, 196], "model": [98], "parallelism,": [99, 101], "pipeline": [100], "memory": [102, 197], "optimization": [103], "strategies,": [104], "design": [107], "sparse": [109], "expert": [111], "models.": [112], "We": [113, 154], "assess": [114], "system-level": [115], "techniques": [116, 166], "such": [117], "ZeRO,": [119], "DeepSpeed,": [120], "tensor": [122], "sharding": [123], "that": [124, 143, 167, 267], "allow": [125], "for": [126, 260, 298], "scalable,": [127], "memory-efficient": [128], "at": [130], "trillion-parameter": [131], "scale.": [132], "Next,": [133], "we": [134, 183, 221], "turn": [135], "to": [136, 146, 209, 252, 283], "multimodality,": [137], "surveying": [138], "architectures": [139], "objectives": [142], "extend": [144], "LLMs": [145], "process": [147], "generate": [149], "diverse": [151], "input": [152], "types.": [153], "review": [155], "contrastive": [156], "learning,": [157], "cross-attention": [158], "fusion,": [159], "aligned": [161], "token": [162], "embeddings": [163], "key": [165, 224], "enable": [168], "cross-modal": [169], "reasoning,": [170, 279], "with": [171], "illustrative": [172], "examples": [173], "from": [174], "like": [176], "Flamingo,": [177], "CLIP,": [178], "GPT-4V.Beyond": [180], "current": [181], "methodologies,": [182], "identify": [184], "formalize": [186], "core": [188], "technical": [189, 293], "challenges": [190], "facing": [191], "LLMs,": [195], "bottlenecks,": [198], "communication": [199], "overhead,": [200], "alignment": [201], "absence": [204], "ground": [206], "truth,": [207], "robustness": [208], "modality": [210], "shifts,": [211], "evaluation": [213], "under": [214], "open-ended": [215], "tasks.": [216], "To": [217], "guide": [218], "future": [219, 304], "research,": [220], "outline": [222], "six": [223], "directions:": [225], "memory-augmented": [227], "architectures,": [228], "modular": [229], "composable": [231], "systems,": [232], "self-aligning": [233], "mechanisms,": [234], "lifelong": [235], "continual": [237], "learning": [238], "agents,": [239], "embodied": [240], "cognition,": [242], "emergence": [245], "general-purpose": [247], "foundation": [248, 265], "agents.Our": [249], "goal": [250], "is": [251], "synthesize": [253], "recent": [254], "progress": [255], "while": [256], "articulating": [257], "vision": [259], "next": [262], "generation": [263], "models\u2014models": [266], "are": [268, 275], "not": [269], "only": [270], "scalable": [271], "but": [274], "also": [276], "capable": [277], "grounding,": [280], "adapting": [282], "complex,": [284], "real-world": [285], "environments.": [286], "serves": [289], "reference": [294], "roadmap": [297], "researchers": [299], "practitioners": [301], "navigating": [302], "large-scale,": [306], "multimodal,": [307], "AI": [310], "systems.": [311]}, "primary_topic": {"id": "https://openalex.org/T11714", "display_name": "Multimodal Machine Learning Applications", "score": 0.9922000169754028, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T11714", "display_name": "Multimodal Machine Learning Applications", "score": 0.9922000169754028, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.004000000189989805, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.0012000000569969416, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.20944/preprints202505.1156.v1", "is_oa": true, "landing_page_url": "https://doi.org/10.20944/preprints202505.1156.v1", "pdf_url": "https://www.preprints.org/frontend/manuscript/2ae29014bca4e5e671379c08ccfdd131/download_pub", "source": {"id": "https://openalex.org/S6309402219", "display_name": "Preprints.org", "issn_l": null, "issn": null, "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": null, "host_organization_name": null, "host_organization_lineage": ["https://openalex.org/P4310310987"], "host_organization_lineage_names": ["Multidisciplinary Digital Publishing Institute"], "type": "repository"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "acceptedVersion", "is_accepted": true, "is_published": false, "raw_source_name": null, "raw_type": "posted-content"}, "cited_by_count": 1, "is_retracted": false}, {"id": "https://openalex.org/W4410468254", "doi": "https://doi.org/10.30574/wjarr.2025.26.2.1793", "title": "Scaling machine learning and operations research models for omni-channel retail in the cloud: A framework for real-time decision optimization", "publication_year": 2025, "publication_date": "2025-05-18", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5117512890", "display_name": "Rakesh Chowdary Ganta", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I39422238", "display_name": "University of Illinois Chicago", "ror": "https://ror.org/02mpq6x41", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I39422238"]}], "countries": ["US"], "is_corresponding": true, "raw_author_name": "Rakesh Chowdary Ganta", "raw_affiliation_strings": ["University of Illinois at Chicago , U. S.A"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "University of Illinois at Chicago , U. S.A", "institution_ids": ["https://openalex.org/I39422238"]}]}], "abstract_inverted_index": {"This": [0], "article": [1], "examines": [2], "how": [3, 32], "cloud-native": [4], "architectures": [5, 110], "enable": [6], "retailers": [7], "to": [8, 27], "scale": [9], "machine": [10], "learning": [11, 97], "and": [12, 45, 62, 95, 103, 115, 126, 138], "operations": [13], "research": [14], "models": [15, 105], "across": [16], "omni-channel": [17], "environments.": [18], "It": [19], "explores": [20], "the": [21, 37, 131], "transformation": [22], "from": [23], "monolithic": [24], "on-premise": [25], "systems": [26], "flexible": [28], "cloud": [29], "platforms,": [30], "highlighting": [31], "distributed": [33, 56], "computing": [34], "frameworks": [35], "address": [36], "computational": [38], "demands": [39], "of": [40, 71, 145], "retail-scale": [41], "ML": [42, 73], "model": [43], "training": [44, 57], "inference.": [46], "The": [47, 69], "discussion": [48], "covers": [49], "architectural": [50, 136], "patterns": [51, 88, 118], "for": [52, 65, 111, 119], "real-time": [53], "data": [54], "processing,": [55], "techniques,": [58], "auto-scaling": [59], "inference": [60], "architectures,": [61], "parallelization": [63], "strategies": [64], "complex": [66], "optimization": [67, 78], "problems.": [68], "integration": [70, 87], "predictive": [72, 102], "insights": [74], "with": [75, 85], "prescriptive": [76, 104], "OR": [77], "is": [79], "presented": [80], "as": [81], "a": [82], "critical": [83], "capability,": [84], "various": [86], "examined": [89], "including": [90], "sequential,": [91], "feedback": [92], "loop,": [93], "stochastic,": [94], "joint": [96], "approaches.": [98], "Data": [99], "pipelines": [100], "connecting": [101], "are": [106], "explored": [107], "alongside": [108], "event-driven": [109], "cross-channel": [112], "decision": [113], "workflows": [114], "API": [116], "design": [117], "unified": [120], "retail": [121, 147], "intelligence": [122], "systems.": [123], "Implementation": [124], "challenges": [125], "technical": [127], "debt": [128], "considerations": [129], "complete": [130], "analysis,": [132], "focusing": [133], "on": [134], "both": [135], "principles": [137], "organizational": [139], "factors": [140], "that": [141], "influence": [142], "successful": [143], "adoption": [144], "cloud-scaled": [146], "analytics": [148]}, "primary_topic": {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.7310000061988831, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.7310000061988831, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11891", "display_name": "Big Data and Business Intelligence", "score": 0.155799999833107, "subfield": {"id": "https://openalex.org/subfields/1404", "display_name": "Management Information Systems"}, "field": {"id": "https://openalex.org/fields/14", "display_name": "Business, Management and Accounting"}, "domain": {"id": "https://openalex.org/domains/2", "display_name": "Social Sciences"}}, {"id": "https://openalex.org/T12885", "display_name": "Digital Innovation in Industries", "score": 0.05620000138878822, "subfield": {"id": "https://openalex.org/subfields/1405", "display_name": "Management of Technology and Innovation"}, "field": {"id": "https://openalex.org/fields/14", "display_name": "Business, Management and Accounting"}, "domain": {"id": "https://openalex.org/domains/2", "display_name": "Social Sciences"}}], "primary_location": {"id": "doi:10.30574/wjarr.2025.26.2.1793", "is_oa": true, "landing_page_url": "https://doi.org/10.30574/wjarr.2025.26.2.1793", "pdf_url": "https://journalwjarr.com/sites/default/files/fulltext_pdf/WJARR-2025-1793.pdf", "source": {"id": "https://openalex.org/S4210223820", "display_name": "World Journal of Advanced Research and Reviews", "issn_l": "2581-9615", "issn": ["2581-9615"], "is_oa": false, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/P4383735917", "host_organization_name": "GSC Online Press", "host_organization_lineage": ["https://openalex.org/P4383735917"], "host_organization_lineage_names": ["GSC Online Press"], "type": "journal"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "World Journal of Advanced Research and Reviews", "raw_type": "journal-article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4416854069", "doi": "https://doi.org/10.48550/arxiv.2508.05387", "title": "Echo: Decoupling Inference and Training for Large-Scale RL Alignment on Heterogeneous Swarms", "publication_year": 2025, "publication_date": "2025-08-07", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5074719547", "display_name": "J. Xiao", "orcid": "https://orcid.org/0000-0002-7860-3958", "observed_orcids": ["https://orcid.org/0000-0002-7860-3958"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Xiao, Jie", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5112209688", "display_name": "Changyuan Fan", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Fan, Changyuan", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": null, "display_name": "Ren, Qingnan", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Ren, Qingnan", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": null, "display_name": "Long, Alfred", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Long, Alfred", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100424305", "display_name": "Yuchen Zhang", "orcid": "https://orcid.org/0000-0001-8536-7175", "observed_orcids": ["https://orcid.org/0000-0001-8536-7175"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Zhang, Yuchen", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": null, "display_name": "Yu, Rymon", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Yu, Rymon", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": null, "display_name": "Yang, Eric", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Yang, Eric", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5119027766", "display_name": "Lynn Ai", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Ai, Lynn", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5112618683", "display_name": "Shaoduo Gan", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Gan, Shaoduo", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Modern": [0], "RL-based": [1], "post-training": [2], "for": [3, 83, 147], "large": [4], "language": [5], "models": [6], "(LLMs)": [7], "co-locate": [8], "trajectory": [9, 134], "sampling": [10], "and": [11, 26, 58, 86, 112, 129], "policy": [12, 77], "optimisation": [13], "on": [14, 114], "the": [15, 20, 34, 46], "same": [16], "GPU": [17], "cluster,": [18, 118], "forcing": [19], "system": [21, 48], "to": [22, 80, 99, 136], "switch": [23], "between": [24], "inference": [25], "training": [27, 41], "workloads.": [28], "This": [29], "serial": [30], "context": [31], "switching": [32], "violates": [33], "single-program-multiple-data": [35], "(SPMD)": [36], "assumption": [37], "underlying": [38], "today's": [39], "distributed": [40, 117], "systems.": [42], "We": [43], "present": [44], "Echo,": [45], "RL": [47, 106, 146], "that": [49, 75, 91, 144], "cleanly": [50], "decouples": [51], "these": [52], "two": [53, 67], "phases": [54], "across": [55], "heterogeneous": [56, 155], "\"inference\"": [57], "\"training\"": [59], "swarms": [60], "while": [61, 132], "preserving": [62], "statistical": [63], "efficiency.": [64], "Echo": [65, 119], "introduces": [66], "lightweight": [68], "synchronization": [69], "protocols:": [70], "a": [71, 96, 115, 121], "sequential": [72], "pull": [73], "mode": [74, 90], "refreshes": [76], "weights": [78], "according": [79], "API": [81], "call": [82], "minimal": [84], "bias,": [85], "an": [87], "asynchronous": [88], "push-pull": [89], "streams": [92], "version-tagged": [93], "rollouts": [94], "through": [95], "replay": [97], "buffer": [98], "maximise": [100], "hardware": [101], "utilisation.": [102], "Training": [103], "four": [104], "representative": [105], "workloads": [107], "with": [108], "Qwen3-4B,": [109], "Qwen2.5-7B,": [110], "Qwen3-30B-A3B-Thinking-2507": [111], "Qwen3-32B": [113], "geographically": [116], "matches": [120], "fully": [122], "co-located": [123], "Verl": [124], "baseline": [125], "in": [126], "convergence": [127], "speed": [128], "final": [130], "reward": [131], "off-loading": [133], "generation": [135], "commodity": [137], "edge": [138], "hardware.": [139], "These": [140], "promising": [141], "results": [142], "demonstrate": [143], "large-scale": [145], "LLMs": [148], "could": [149], "achieve": [150], "datacentre-grade": [151], "performance": [152], "using": [153], "decentralised,": [154], "resources.": [156]}, "primary_topic": {"id": "https://openalex.org/T10462", "display_name": "Reinforcement Learning in Robotics", "score": 0.2808000147342682, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10462", "display_name": "Reinforcement Learning in Robotics", "score": 0.2808000147342682, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.19480000436306, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10181", "display_name": "Natural Language Processing Techniques", "score": 0.06300000101327896, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2508.05387", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2508.05387", "pdf_url": "https://arxiv.org/pdf/2508.05387", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4407719187", "doi": "https://doi.org/10.35629/5252-0702382390", "title": "Scalable AI: Leveraging Cloud Infrastructure for Large-Scale Machine Learning", "publication_year": 2025, "publication_date": "2025-02-01", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5015470915", "display_name": "Abdul Muqtadir Mohammed", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I63190737", "display_name": "University at Buffalo, State University of New York", "ror": "https://ror.org/01y64my43", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I63190737"]}], "countries": ["US"], "is_corresponding": true, "raw_author_name": "Abdul Muqtadir Mohammed", "raw_affiliation_strings": ["University at Buffalo , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "University at Buffalo , USA", "institution_ids": ["https://openalex.org/I63190737"]}]}], "abstract_inverted_index": {"This": [0], "comprehensive": [1], "article": [2, 17, 50, 75], "explores": [3], "the": [4, 37], "evolution": [5], "and": [6, 28, 47, 60, 70, 82], "implementation": [7], "of": [8, 21, 39], "cloud": [9, 33], "infrastructure": [10, 55], "for": [11, 66, 95], "large-scale": [12], "machine": [13, 87], "learning": [14, 88], "systems.": [15], "The": [16, 49, 74], "examines": [18], "critical": [19], "aspects": [20], "distributed": [22], "training": [23, 72], "architectures,": [24, 62], "resource": [25], "optimization": [26], "strategies,": [27], "performance": [29], "enhancement": [30], "techniques": [31], "in": [32, 85], "environments.": [34], "It": [35], "addresses": [36], "challenges": [38], "scaling": [40], "artificial": [41], "intelligence": [42], "workloads": [43], "while": [44], "maintaining": [45], "efficiency": [46], "cost-effectiveness.": [48], "analyzes": [51], "various": [52], "approaches": [53], "to": [54], "design,": [56], "including": [57], "parameter": [58], "server": [59], "ringallreduce": [61], "along": [63], "with": [64], "methods": [65], "optimizing": [67], "data": [68], "pipelines": [69], "model": [71], "processes.": [73], "also": [76], "covers": [77], "monitoring": [78], "systems,": [79], "security": [80], "considerations,": [81], "emerging": [83], "trends": [84], "cloud-based": [86], "deployments,": [89], "providing": [90], "insights": [91], "into": [92], "best": [93], "practices": [94], "implementing": [96], "scalable": [97], "AI": [98], "solutions.": [99]}, "primary_topic": {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.9990000128746033, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.9990000128746033, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.00019999999494757503, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.00019999999494757503, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.35629/5252-0702382390", "is_oa": true, "landing_page_url": "https://doi.org/10.35629/5252-0702382390", "pdf_url": "https://doi.org/10.35629/5252-0702382390", "source": {"id": "https://openalex.org/S5406996880", "display_name": "International Journal of Advances in Engineering and Management", "issn_l": "2395-5252", "issn": ["2395-5252"], "is_oa": false, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": null, "host_organization_name": null, "host_organization_lineage": [], "host_organization_lineage_names": [], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "International Journal of Advances in Engineering and Management", "raw_type": "journal-article"}, "cited_by_count": 1, "is_retracted": false}, {"id": "https://openalex.org/W4417308080", "doi": "https://doi.org/10.48550/arxiv.2506.09280", "title": "TTrace: Lightweight Error Checking and Diagnosis for Distributed Training", "publication_year": 2025, "publication_date": "2025-06-10", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5017537432", "display_name": "Haitian Jiang", "orcid": "https://orcid.org/0000-0001-8215-166X", "observed_orcids": ["https://orcid.org/0000-0001-8215-166X", "https://orcid.org/0009-0009-9331-4773"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Jiang, Haitian", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5103252880", "display_name": "Shaowei Zhu", "orcid": "https://orcid.org/0000-0002-2224-7069", "observed_orcids": ["https://orcid.org/0000-0002-2224-7069", "https://orcid.org/0009-0000-3893-2231"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Zhu, Shaowei", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100704412", "display_name": "Zhen Zhang", "orcid": "https://orcid.org/0000-0003-2836-1572", "observed_orcids": ["https://orcid.org/0000-0003-2836-1572"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Zhang, Zhen", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5102001023", "display_name": "Zhenyu Song", "orcid": "https://orcid.org/0000-0003-1548-1858", "observed_orcids": ["https://orcid.org/0000-0003-1548-1858"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Song, Zhenyu", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5020914269", "display_name": "Xinwei Fu", "orcid": "https://orcid.org/0009-0004-7822-5450", "observed_orcids": ["https://orcid.org/0009-0004-7822-5450"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Fu, Xinwei", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5023640358", "display_name": "Zhen Jia", "orcid": "https://orcid.org/0000-0002-6810-2279", "observed_orcids": ["https://orcid.org/0000-0002-6810-2279"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Jia, Zhen", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5101638214", "display_name": "Yida Wang", "orcid": "https://orcid.org/0000-0001-8165-840X", "observed_orcids": ["https://orcid.org/0000-0001-8165-840X"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Wang, Yida", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5100320436", "display_name": "Jinyang Li", "orcid": "https://orcid.org/0009-0000-6057-8937", "observed_orcids": ["https://orcid.org/0009-0000-6057-8937"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Li, Jinyang", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Distributed": [0], "training": [1, 7, 28, 48, 59, 68, 114, 188, 201], "is": [2, 60, 184], "essential": [3], "for": [4, 99, 142], "scaling": [5], "the": [6, 24, 93, 125, 129, 169, 206], "of": [8, 21, 26, 180], "large": [9, 15], "neural": [10], "network": [11], "models,": [12], "such": [13, 54], "as": [14], "language": [16], "models": [17], "(LLMs),": [18], "across": [19], "thousands": [20], "GPUs.": [22], "However,": [23], "complexity": [25], "distributed": [27, 58, 106, 113], "programs": [29], "makes": [30], "them": [31], "particularly": [32], "prone": [33], "to": [34, 46, 81, 147], "silent": [35, 55, 103], "bugs,": [36], "which": [37], "do": [38], "not": [39], "produce": [40], "explicit": [41], "error": [42], "signals": [43], "but": [44], "lead": [45], "incorrect": [47], "outcomes.": [49], "Effectively": [50], "detecting": [51, 100], "and": [52, 77, 90, 101, 164, 195], "localizing": [53, 102], "bugs": [56, 104, 163, 167], "in": [57, 105, 128, 168, 186, 211], "challenging.": [61], "Common": [62], "debugging": [63], "practices": [64], "based": [65], "on": [66], "monitoring": [67], "loss": [69], "or": [70], "gradient": [71], "norm": [72], "curves": [73], "are": [74], "indirect,": [75], "inefficient,": [76], "provide": [78], "no": [79], "way": [80], "localize": [82], "bugs.": [83], "To": [84, 122], "address": [85], "those": [86, 116], "challenges,": [87], "we": [88, 132], "design": [89], "implement": [91], "TTrace,": [92], "first": [94], "systematic": [95], "differential": [96], "testing": [97], "system": [98], "training.": [107], "TTrace": [108, 146, 158, 183, 210], "aligns": [109], "intermediate": [110], "tensors": [111], "from": [112, 117, 151], "with": [115], "a": [118, 134, 140, 198], "trusted": [119], "reference": [120], "implementation.": [121], "properly": [123], "compare": [124], "floating-point": [126], "values": [127], "corresponding": [130], "tensors,": [131], "propose": [133], "novel": [135], "mathematical": [136], "analysis": [137], "that": [138, 157], "provides": [139], "guideline": [141], "setting": [143], "tolerances,": [144], "enabling": [145], "distinguish": [148], "bug-induced": [149], "errors": [150], "numerical": [152], "errors.": [153], "Experimental": [154], "results": [155], "demonstrate": [156], "effectively": [159], "detects": [160], "11": [161], "existing": [162], "3": [165], "new": [166], "widely": [170], "used": [171], "Megatron-LM": [172], "framework,": [173], "while": [174], "requiring": [175], "fewer": [176], "than": [177], "10": [178], "lines": [179], "code": [181], "changes.": [182], "effective": [185], "various": [187], "recipes,": [189], "including": [190], "low-precision": [191], "recipes": [192], "involving": [193], "BF16": [194], "FP8.": [196], "Notably,": [197], "popular": [199], "open-source": [200], "framework": [202], "has": [203], "already": [204], "adopted": [205], "method": [207], "proposed": [208], "by": [209], "its": [212], "development": [213], "workflow.": [214]}, "primary_topic": {"id": "https://openalex.org/T10743", "display_name": "Software Testing and Debugging Techniques", "score": 1.0, "subfield": {"id": "https://openalex.org/subfields/1712", "display_name": "Software"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10743", "display_name": "Software Testing and Debugging Techniques", "score": 1.0, "subfield": {"id": "https://openalex.org/subfields/1712", "display_name": "Software"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T12127", "display_name": "Software System Performance and Reliability", "score": 9.999999747378752e-05, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2506.09280", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2506.09280", "pdf_url": "https://arxiv.org/pdf/2506.09280", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4407774982", "doi": "https://doi.org/10.62051/ijcsit.v5n1.05", "title": "Practice and Optimization of Deep Learning Model Training", "publication_year": 2025, "publication_date": "2025-01-23", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5108047153", "display_name": "Yiwei Dong", "orcid": "https://orcid.org/0000-0002-8698-3962", "observed_orcids": ["https://orcid.org/0000-0002-8698-3962", "https://orcid.org/0009-0000-5512-9109"]}, "institutions": [{"id": "https://openalex.org/I25355098", "display_name": "Chang'an University", "ror": "https://ror.org/05mxya461", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I25355098"]}], "countries": ["CN"], "is_corresponding": true, "raw_author_name": "Yiwei Dong", "raw_affiliation_strings": ["School of Information Engineering , Chang'an University , Xi'an , Shaanxi , China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Information Engineering , Chang'an University , Xi'an , Shaanxi , China", "institution_ids": ["https://openalex.org/I25355098"]}]}], "abstract_inverted_index": {"As": [0], "IT": [1], "rapidly": [2], "advances,": [3], "deep": [4, 50, 186, 208], "learning,": [5, 51, 131], "a": [6, 118, 212], "key": [7], "AI": [8], "area,": [9], "has": [10], "achieved": [11], "significant": [12], "results": [13], "in": [14, 85, 197], "image": [15], "recognition,": [16], "NLP,": [17], "and": [18, 26, 36, 47, 58, 62, 94, 106, 140, 154, 169, 192, 200, 211], "speech": [19], "recognition.": [20], "This": [21, 41, 183], "research": [22], "focuses": [23], "on": [24], "\"Practice": [25], "Optimization": [27], "of": [28, 49, 66, 88, 114, 120, 151, 173], "Deep": [29], "Learning": [30], "Model": [31], "Training,\"": [32], "detailing": [33], "efficient": [34, 121], "methods": [35], "techniques": [37], "for": [38, 166, 195, 207, 215], "practical": [39, 205], "applications.": [40], "paper": [42, 184], "reviews": [43, 185], "the": [44, 64, 74, 86, 95, 112, 115, 148, 152, 156, 170, 174], "basic": [45], "theory": [46], "evolution": [48], "summarizes": [52], "some": [53, 161], "mainstream": [54], "neural": [55], "network": [56], "architectures": [57], "their": [59], "applicable": [60], "scenarios,": [61], "emphasizes": [63], "importance": [65], "data": [67, 127], "preprocessing": [68], "to": [69], "improve": [70], "model": [71, 89, 153], "performance.": [72, 158, 202], "Then,": [73], "common": [75], "problems": [76, 96], "such": [77, 97], "as": [78, 98], "overfitting,": [79], "underfitting,": [80], "gradient": [81], "disappearing": [82], "or": [83], "explosion": [84], "process": [87], "training": [90, 139], "are": [91, 109, 124, 164], "deeply": [92], "analyzed,": [93], "loss": [99], "function": [100], "selection,": [101], "backpropagation": [102], "mechanism,": [103], "optimizer": [104], "configuration": [105], "hyperparameter": [107], "adjustment": [108], "discussed.": [110], "In": [111, 159], "face": [113], "above": [116], "challenges,": [117], "series": [119], "optimization": [122, 176, 196], "schemes": [123], "proposed,": [125], "including": [126], "enhancement": [128], "techniques,": [129], "transfer": [130], "regularization": [132], "methods,": [133], "adaptive": [134], "learning": [135, 187, 209], "rate": [136], "adjustment,": [137], "distributed": [138], "mixed": [141], "precision": [142], "training,": [143], "etc.,": [144], "which": [145], "can": [146], "accelerate": [147], "convergence": [149], "speed": [150], "enhance": [155], "final": [157], "addition,": [160], "concrete": [162], "cases": [163], "selected": [165], "empirical": [167], "study,": [168], "actual": [171], "effect": [172], "proposed": [175], "scheme": [177], "is": [178], "tested": [179], "by": [180], "comparative": [181], "experiments.": [182], "research,": [188], "discusses": [189], "future": [190], "directions,": [191], "notes": [193], "room": [194], "theoretical": [198], "architecture": [199], "computational": [201], "It": [203], "offers": [204], "guidance": [206], "experts": [210], "new": [213], "perspective": [214], "related": [216], "scholars.": [217]}, "primary_topic": {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.32839998602867126, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.32839998602867126, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.22439999878406525, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T12535", "display_name": "Machine Learning and Data Classification", "score": 0.1965000033378601, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.62051/ijcsit.v5n1.05", "is_oa": true, "landing_page_url": "https://doi.org/10.62051/ijcsit.v5n1.05", "pdf_url": "https://wepub.org/index.php/IJCSIT/article/download/5068/5610", "source": {"id": "https://openalex.org/S4390963318", "display_name": "International Journal of Computer Science and Information Technology", "issn_l": "3005-7140", "issn": ["3005-7140", "3005-9682"], "is_oa": true, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core"], "host_organization": null, "host_organization_name": null, "host_organization_lineage": [], "host_organization_lineage_names": [], "type": "journal"}, "license": "cc-by-nc", "license_id": "https://openalex.org/licenses/cc-by-nc", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "International Journal of Computer Science and Information Technology", "raw_type": "journal-article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W7114788792", "doi": "https://doi.org/10.48550/arxiv.2512.08242", "title": "Chopper: A Multi-Level GPU Characterization Tool & Derived Insights Into LLM Training Inefficiency", "publication_year": 2025, "publication_date": "2025-12-09", "type": "preprint", "language": null, "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Kurzynski, Marco", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I106165777", "display_name": "University of Central Florida", "ror": "https://ror.org/036nfer12", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I106165777"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Kurzynski, Marco", "raw_affiliation_strings": ["University of Central Florida"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "University of Central Florida", "institution_ids": ["https://openalex.org/I106165777"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Aga, Shaizeen", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I4210137977", "display_name": "Advanced Micro Devices (United States)", "ror": "https://ror.org/04kd6c783", "country_code": "US", "type": "company", "lineage": ["https://openalex.org/I4210137977"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Aga, Shaizeen", "raw_affiliation_strings": ["Advanced Micro Devices , Inc"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Advanced Micro Devices , Inc", "institution_ids": ["https://openalex.org/I4210137977"]}]}, {"author_position": "last", "author": {"id": null, "display_name": "Wu, Di", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I106165777", "display_name": "University of Central Florida", "ror": "https://ror.org/036nfer12", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I106165777"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Wu, Di", "raw_affiliation_strings": ["University of Central Florida"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "University of Central Florida", "institution_ids": ["https://openalex.org/I106165777"]}]}], "abstract_inverted_index": {"Training": [0], "large": [1], "language": [2], "models": [3], "(LLMs)": [4], "efficiently": [5], "requires": [6], "a": [7, 57, 92], "deep": [8], "understanding": [9], "of": [10, 96, 139, 162, 180], "how": [11], "modern": [12], "GPU": [13, 67, 131, 201], "systems": [14], "behave": [15], "under": [16, 101], "real-world": [17], "distributed": [18], "training": [19, 47, 100, 182, 193], "workloads.": [20], "While": [21], "prior": [22], "work": [23], "has": [24], "focused": [25], "primarily": [26], "on": [27, 107, 183], "kernel-level": [28], "performance": [29, 72], "or": [30], "single-GPU": [31], "microbenchmarks,": [32], "the": [33, 147, 152, 160, 175], "complex": [34], "interaction": [35], "between": [36, 154], "communication,": [37], "computation,": [38], "memory": [39, 125, 133], "behavior,": [40], "and": [41, 59, 65, 70, 86, 121, 132, 156, 168, 198, 203], "power": [42], "management": [43], "in": [44], "multi-GPU": [45], "LLM": [46, 181], "remains": [48], "poorly": [49], "characterized.": [50], "In": [51], "this": [52], "work,": [53], "we": [54, 90], "introduce": [55], "Chopper,": [56, 89], "profiling": [58], "analysis": [60, 115], "framework": [61], "that": [62], "collects,": [63], "aligns,": [64], "visualizes": [66], "kernel": [68, 169], "traces": [69], "hardware": [71], "counters": [73], "across": [74], "multiple": [75], "granularities": [76], "(i.e.,": [77], "from": [78], "individual": [79], "kernels": [80], "to": [81, 151], "operations,": [82], "layers,": [83], "phases,": [84], "iterations,": [85], "GPUs).": [87], "Using": [88], "perform": [91], "comprehensive": [93], "end-to-end": [94], "characterization": [95, 179], "Llama": [97], "3": [98], "8B": [99], "fully": [102], "sharded": [103], "data": [104], "parallelism": [105], "(FSDP)": [106], "an": [108], "eight-GPU": [109], "AMD": [110, 184], "InstinctTM": [111, 185], "MI300X": [112, 186], "node.": [113], "Our": [114], "reveals": [116], "several": [117, 137], "previously": [118], "underexplored": [119], "bottlenecks": [120], "behaviors,": [122], "such": [123], "as": [124], "determinism": [126], "enabling": [127], "higher,": [128], "more": [129], "stable": [130], "frequencies.": [134], "We": [135], "identify": [136], "sources": [138], "inefficiencies,": [140], "with": [141], "frequency": [142], "overhead": [143], "(DVFS": [144], "effects)": [145], "being": [146], "single": [148], "largest": [149], "contributor": [150], "gap": [153], "theoretical": [155], "observed": [157], "performance,": [158], "exceeding": [159], "impact": [161], "MFMA": [163], "utilization": [164], "loss,": [165], "communication/computation": [166], "overlap,": [167], "launch": [170], "overheads.": [171], "Overall,": [172], "Chopper": [173], "provides": [174], "first": [176], "holistic,": [177], "multi-granularity": [178], "GPUs,": [187], "yielding": [188], "actionable": [189], "insights": [190], "for": [191], "optimizing": [192], "frameworks,": [194], "improving": [195], "power-management": [196], "strategies,": [197], "guiding": [199], "future": [200], "architecture": [202], "system": [204], "design.": [205]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9994999766349792, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9994999766349792, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T12127", "display_name": "Software System Performance and Reliability", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10481", "display_name": "Computer Graphics and Visualization Techniques", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/1704", "display_name": "Computer Graphics and Computer-Aided Design"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2512.08242", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2512.08242", "pdf_url": "https://arxiv.org/pdf/2512.08242", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4410536548", "doi": "https://doi.org/10.1109/access.2025.3571979", "title": "A Framework Integrating Federated Learning and Fog Computing Based on Client Sampling and Dynamic Thresholding Techniques", "publication_year": 2025, "publication_date": "2025-01-01", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5029758324", "display_name": "Dang Van Thang", "orcid": "https://orcid.org/0009-0009-2219-3767", "observed_orcids": ["https://orcid.org/0009-0009-2219-3767"]}, "institutions": [{"id": "https://openalex.org/I37355250", "display_name": "Saint-Petersburg State University of Telecommunications", "ror": "https://ror.org/00pcyc255", "country_code": "RU", "type": "education", "lineage": ["https://openalex.org/I37355250"]}], "countries": ["RU"], "is_corresponding": false, "raw_author_name": "Dang van Thang", "raw_affiliation_strings": ["Department of Telecommunication Networks and Data Transmission, The Bonch-Bruevich Saint-Petersburg State University of Telecommunications, Saint Petersburg, Russia"], "raw_orcid": "https://orcid.org/0009-0009-2219-3767", "affiliations": [{"raw_affiliation_string": "Department of Telecommunication Networks and Data Transmission, The Bonch-Bruevich Saint-Petersburg State University of Telecommunications, Saint Petersburg, Russia", "institution_ids": ["https://openalex.org/I37355250"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5053245403", "display_name": "Artem Volkov", "orcid": "https://orcid.org/0009-0002-4296-1822", "observed_orcids": ["https://orcid.org/0009-0002-4296-1822"]}, "institutions": [{"id": "https://openalex.org/I37355250", "display_name": "Saint-Petersburg State University of Telecommunications", "ror": "https://ror.org/00pcyc255", "country_code": "RU", "type": "education", "lineage": ["https://openalex.org/I37355250"]}], "countries": ["RU"], "is_corresponding": false, "raw_author_name": "Artem Volkov", "raw_affiliation_strings": ["Department of Telecommunication Networks and Data Transmission, The Bonch-Bruevich Saint-Petersburg State University of Telecommunications, Saint Petersburg, Russia"], "raw_orcid": "https://orcid.org/0009-0002-4296-1822", "affiliations": [{"raw_affiliation_string": "Department of Telecommunication Networks and Data Transmission, The Bonch-Bruevich Saint-Petersburg State University of Telecommunications, Saint Petersburg, Russia", "institution_ids": ["https://openalex.org/I37355250"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5024357714", "display_name": "Ammar Muthanna", "orcid": "https://orcid.org/0000-0003-0213-8145", "observed_orcids": ["https://orcid.org/0000-0003-0213-8145", "https://orcid.org/0000-0002-1069-3993"]}, "institutions": [{"id": "https://openalex.org/I37355250", "display_name": "Saint-Petersburg State University of Telecommunications", "ror": "https://ror.org/00pcyc255", "country_code": "RU", "type": "education", "lineage": ["https://openalex.org/I37355250"]}], "countries": ["RU"], "is_corresponding": false, "raw_author_name": "Ammar Muthanna", "raw_affiliation_strings": ["Department of Telecommunication Networks and Data Transmission, The Bonch-Bruevich Saint-Petersburg State University of Telecommunications, Saint Petersburg, Russia"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Telecommunication Networks and Data Transmission, The Bonch-Bruevich Saint-Petersburg State University of Telecommunications, Saint Petersburg, Russia", "institution_ids": ["https://openalex.org/I37355250"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5030075164", "display_name": "Ibrahim A. Elgendy", "orcid": "https://orcid.org/0000-0001-7154-2307", "observed_orcids": ["https://orcid.org/0000-0001-7154-2307"]}, "institutions": [{"id": "https://openalex.org/I134085113", "display_name": "King Fahd University of Petroleum and Minerals", "ror": "https://ror.org/03yez3163", "country_code": "SA", "type": "education", "lineage": ["https://openalex.org/I134085113"]}], "countries": ["SA"], "is_corresponding": false, "raw_author_name": "Ibrahim A. Elgendy", "raw_affiliation_strings": ["IRC for Finance and Digital Economy, KFUPM Business School, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia"], "raw_orcid": "https://orcid.org/0000-0001-7154-2307", "affiliations": [{"raw_affiliation_string": "IRC for Finance and Digital Economy, KFUPM Business School, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia", "institution_ids": ["https://openalex.org/I134085113"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5020065096", "display_name": "Reem Ibrahim Alkanhel", "orcid": "https://orcid.org/0000-0001-6395-4723", "observed_orcids": ["https://orcid.org/0000-0001-6395-4723"]}, "institutions": [{"id": "https://openalex.org/I106778892", "display_name": "Princess Nourah bint Abdulrahman University", "ror": "https://ror.org/05b0cyh02", "country_code": "SA", "type": "education", "lineage": ["https://openalex.org/I106778892"]}], "countries": ["SA"], "is_corresponding": false, "raw_author_name": "Reem Alkanhel", "raw_affiliation_strings": ["Department of Information Technology, College of Computer and Information Sciences, Princess Nourah bint Abdulrahman University, P.O. Box 84428, Riyadh, Saudi Arabia"], "raw_orcid": "https://orcid.org/0000-0001-6395-4723", "affiliations": [{"raw_affiliation_string": "Department of Information Technology, College of Computer and Information Sciences, Princess Nourah bint Abdulrahman University, P.O. Box 84428, Riyadh, Saudi Arabia", "institution_ids": ["https://openalex.org/I106778892"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5062076081", "display_name": "Dushantha Nalin K. Jayakody", "orcid": "https://orcid.org/0000-0002-7004-2930", "observed_orcids": ["https://orcid.org/0000-0002-7004-2930"]}, "institutions": [{"id": "https://openalex.org/I4210119063", "display_name": "Autonomous University of Lisbon", "ror": "https://ror.org/01ryrwk91", "country_code": "PT", "type": "education", "lineage": ["https://openalex.org/I4210119063"]}, {"id": "https://openalex.org/I88596544", "display_name": "Universidade Lus\u00f3fona", "ror": "https://ror.org/05xxfer42", "country_code": "PT", "type": "education", "lineage": ["https://openalex.org/I88596544"]}], "countries": ["PT"], "is_corresponding": false, "raw_author_name": "Dushantha Nalin K. Jayakody", "raw_affiliation_strings": ["COPELABS, Universidade Lus&#x00F3;fona, Lisbon, Portugal", "Centro de Investiga&#x00E7;&#x00E4; em Tecnologias&#x2014;Aut&#x00F3;noma TechLab, Universidade Aut&#x00F3;noma de Lisboa, Lisbon, PORTUGAL"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "COPELABS, Universidade Lus&#x00F3;fona, Lisbon, Portugal", "institution_ids": ["https://openalex.org/I88596544"]}, {"raw_affiliation_string": "Centro de Investiga&#x00E7;&#x00E4; em Tecnologias&#x2014;Aut&#x00F3;noma TechLab, Universidade Aut&#x00F3;noma de Lisboa, Lisbon, PORTUGAL", "institution_ids": ["https://openalex.org/I4210119063"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5018187110", "display_name": "Andrey Koucheryavy", "orcid": "https://orcid.org/0000-0003-4479-2479", "observed_orcids": ["https://orcid.org/0000-0003-4479-2479"]}, "institutions": [{"id": "https://openalex.org/I37355250", "display_name": "Saint-Petersburg State University of Telecommunications", "ror": "https://ror.org/00pcyc255", "country_code": "RU", "type": "education", "lineage": ["https://openalex.org/I37355250"]}], "countries": ["RU"], "is_corresponding": false, "raw_author_name": "Andrey Koucheryavy", "raw_affiliation_strings": ["Department of Telecommunication Networks and Data Transmission, The Bonch-Bruevich Saint-Petersburg State University of Telecommunications, Saint Petersburg, Russia"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Telecommunication Networks and Data Transmission, The Bonch-Bruevich Saint-Petersburg State University of Telecommunications, Saint Petersburg, Russia", "institution_ids": ["https://openalex.org/I37355250"]}]}], "abstract_inverted_index": {"The": [0], "exponential": [1], "growth": [2], "in": [3, 157], "the": [4, 13, 25, 61, 67, 82, 86, 97, 101, 112, 118, 193, 215], "number": [5], "of": [6, 8, 16, 27, 69, 88, 100, 103], "Internet": [7], "Things": [9], "(IoT)": [10], "devices": [11], "and": [12, 45, 65, 77, 134, 139, 146, 150, 197, 213, 222, 233], "vast": [14], "quantity": [15], "data": [17, 49, 128, 149], "they": [18], "generate": [19], "present": [20], "a": [21, 36, 55, 93, 123, 165], "significant": [22], "challenge": [23], "to": [24, 63, 84, 110, 143], "efficacy": [26], "traditional": [28, 89], "centralized": [29, 90], "training": [30, 122, 127], "models.": [31], "Federated": [32, 173], "Learning": [33, 174], "(FL)": [34], "is": [35, 109], "machine": [37], "learning": [38], "framework": [39, 108, 169], "that": [40, 170, 225], "effectively": [41], "addresses": [42], "this": [43, 105, 161], "issue": [44], "other": [46], "concerns": [47], "about": [48], "privacy.": [50], "Furthermore,": [51], "fog": [52, 78, 199], "computing": [53, 58, 79], "represents": [54], "robust": [56], "distributed": [57, 73, 126], "methodology": [59], "with": [60, 175], "potential": [62, 83], "bolster": [64], "propel": [66], "advancement": [68], "FL.": [70], "An": [71], "integrated": [72], "architecture": [74], "combining": [75], "FL": [76], "(FC)": [80], "has": [81], "overcome": [85], "limitations": [87], "architectures,": [91], "offering": [92], "promising": [94], "solution": [95], "for": [96, 201], "future.": [98], "One": [99], "objectives": [102], "implementing": [104], "novel": [106, 166], "architectural": [107], "alleviate": [111], "burden": [113], "on": [114, 125, 184, 220], "communication": [115, 230], "links": [116], "within": [117], "core": [119], "network": [120], "by": [121, 231], "model": [124, 144], "across": [129], "many": [130], "clients.": [131], "Various": [132], "techniques": [133], "frameworks": [135], "have": [136, 154], "been": [137], "developed": [138], "implemented,": [140], "including": [141], "approaches": [142], "compression": [145], "those": [147], "addressing": [148], "device": [151], "heterogeneity.": [152], "These": [153], "demonstrated": [155], "effectiveness": [156], "specific": [158], "contexts.": [159], "In": [160], "paper,": [162], "we": [163], "introduce": [164], "gradient-driven": [167], "client-sampling": [168], "tightly": [171], "couples": [172], "Fog": [176], "Computing.": [177], "By": [178], "dynamically": [179], "adjusting": [180], "per-round": [181], "thresholds": [182], "based": [183], "local": [185], "gradient": [186], "change": [187], "rates,": [188], "our": [189, 226], "method": [190], "selects": [191], "only": [192], "most": [194], "informative": [195], "clients": [196], "leverages": [198], "nodes": [200], "partial": [202], "aggregation,": [203], "thereby": [204], "minimizing": [205], "redundant": [206], "transmissions,": [207], "accelerating": [208], "convergence": [209, 238], "under": [210], "heterogeneous": [211], "data,": [212], "offloading": [214], "central": [216], "server.": [217], "Extensive": [218], "simulations": [219], "MNIST": [221], "CIFAR-10": [223], "demonstrate": [224], "approach": [227], "reduces": [228], "cumulative": [229], "39%": [232], "31%,": [234], "respectively,": [235], "without": [236], "sacrificing": [237], "speed": [239], "or": [240], "final": [241], "accuracy.": [242]}, "primary_topic": {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.9961000084877014, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.9961000084877014, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.003100000089034438, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 9.999999747378752e-05, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/access.2025.3571979", "is_oa": true, "landing_page_url": "https://doi.org/10.1109/access.2025.3571979", "pdf_url": null, "source": {"id": "https://openalex.org/S2485537415", "display_name": "IEEE Access", "issn_l": "2169-3536", "issn": ["2169-3536"], "is_oa": true, "is_in_doaj": true, "is_core": true, "listed_in": ["cwts-core", "doaj", "jufo-1", "ki-jl-1", "norway-1"], "host_organization": "https://openalex.org/P4310319808", "host_organization_name": "Institute of Electrical and Electronics Engineers", "host_organization_lineage": ["https://openalex.org/P4310319808"], "host_organization_lineage_names": ["Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Access", "raw_type": "journal-article"}, "cited_by_count": 4, "is_retracted": false}, {"id": "https://openalex.org/W4413267660", "doi": "https://doi.org/10.1109/tcad.2025.3593436", "title": "An Elastic Federated Learning Collaboration Framework for Computing-Constrained IoT", "publication_year": 2025, "publication_date": "2025-07-28", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Haizhou Wang", "orcid": "https://orcid.org/0009-0009-1051-2214", "observed_orcids": ["https://orcid.org/0009-0009-1051-2214"]}, "institutions": [{"id": "https://openalex.org/I113940042", "display_name": "Shanghai University", "ror": "https://ror.org/006teas31", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I113940042"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Haizhou Wang", "raw_affiliation_strings": ["School of Computer Engineering and Science, Shanghai University, Shanghai, China"], "raw_orcid": "https://orcid.org/0009-0009-1051-2214", "affiliations": [{"raw_affiliation_string": "School of Computer Engineering and Science, Shanghai University, Shanghai, China", "institution_ids": ["https://openalex.org/I113940042"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5025230449", "display_name": "Guobing Zou", "orcid": "https://orcid.org/0000-0002-7865-8158", "observed_orcids": ["https://orcid.org/0000-0002-7865-8158"]}, "institutions": [{"id": "https://openalex.org/I113940042", "display_name": "Shanghai University", "ror": "https://ror.org/006teas31", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I113940042"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Guobing Zou", "raw_affiliation_strings": ["School of Computer Engineering and Science, Shanghai University, Shanghai, China"], "raw_orcid": "https://orcid.org/0000-0002-7865-8158", "affiliations": [{"raw_affiliation_string": "School of Computer Engineering and Science, Shanghai University, Shanghai, China", "institution_ids": ["https://openalex.org/I113940042"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5052730043", "display_name": "Kun Cao", "orcid": "https://orcid.org/0000-0003-4872-4908", "observed_orcids": ["https://orcid.org/0000-0003-4872-4908"]}, "institutions": [{"id": "https://openalex.org/I159948400", "display_name": "Jinan University", "ror": "https://ror.org/02xe5ns62", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I159948400"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Kun Cao", "raw_affiliation_strings": ["College of Cyber Security, Jinan University, Guangzhou, China"], "raw_orcid": "https://orcid.org/0000-0003-4872-4908", "affiliations": [{"raw_affiliation_string": "College of Cyber Security, Jinan University, Guangzhou, China", "institution_ids": ["https://openalex.org/I159948400"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5028987002", "display_name": "Yangguang Cui", "orcid": "https://orcid.org/0009-0008-0411-6461", "observed_orcids": ["https://orcid.org/0009-0008-0411-6461", "https://orcid.org/0000-0001-9447-0983"]}, "institutions": [{"id": "https://openalex.org/I113940042", "display_name": "Shanghai University", "ror": "https://ror.org/006teas31", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I113940042"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yangguang Cui", "raw_affiliation_strings": ["School of Computer Engineering and Science, Shanghai University, Shanghai, China"], "raw_orcid": "https://orcid.org/0009-0008-0411-6461", "affiliations": [{"raw_affiliation_string": "School of Computer Engineering and Science, Shanghai University, Shanghai, China", "institution_ids": ["https://openalex.org/I113940042"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5042163232", "display_name": "Tongquan Wei", "orcid": "https://orcid.org/0000-0002-7421-1711", "observed_orcids": ["https://orcid.org/0000-0002-7421-1711", "https://orcid.org/0000-0003-4290-3163"]}, "institutions": [{"id": "https://openalex.org/I66867065", "display_name": "East China Normal University", "ror": "https://ror.org/02n96ep67", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I66867065"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Tongquan Wei", "raw_affiliation_strings": ["School of Computer Science and Technology, East China Normal University, Shanghai, China"], "raw_orcid": "https://orcid.org/0000-0002-7421-1711", "affiliations": [{"raw_affiliation_string": "School of Computer Science and Technology, East China Normal University, Shanghai, China", "institution_ids": ["https://openalex.org/I66867065"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5031580282", "display_name": "Shiyan Hu", "orcid": "https://orcid.org/0000-0003-2512-0634", "observed_orcids": ["https://orcid.org/0000-0003-2512-0634"]}, "institutions": [{"id": "https://openalex.org/I889458895", "display_name": "University of Hong Kong", "ror": "https://ror.org/02zhqgq86", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I889458895"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Shiyan Hu", "raw_affiliation_strings": ["Department of Data and Systems Engineering, University of Hong Kong, Pok Fu Lam, SAR, Hong Kong", "Department of Data and Systems Engineering, University of Hong Kong, Hong Kong, Hong Kong"], "raw_orcid": "https://orcid.org/0000-0003-2512-0634", "affiliations": [{"raw_affiliation_string": "Department of Data and Systems Engineering, University of Hong Kong, Pok Fu Lam, SAR, Hong Kong", "institution_ids": ["https://openalex.org/I889458895"]}, {"raw_affiliation_string": "Department of Data and Systems Engineering, University of Hong Kong, Hong Kong, Hong Kong", "institution_ids": ["https://openalex.org/I889458895"]}]}], "abstract_inverted_index": {"Through": [0], "exploiting": [1], "decentralized": [2], "data": [3], "from": [4], "multi-source": [5], "Internet-of-Things": [6], "(IoT)": [7], "devices,": [8, 61], "federated": [9, 76], "learning": [10, 77], "(FL)": [11], "can": [12, 170], "accomplish": [13], "the": [14], "training": [15, 46, 129, 136], "of": [16, 59, 87, 120, 156], "deep": [17], "neural": [18, 97], "network": [19, 98], "(DNN)": [20], "models": [21], "in": [22, 47, 184, 193], "a": [23, 126], "privacy-preserving": [24], "manner": [25], "to": [26, 32, 55, 82, 112, 131, 152, 165, 173, 181, 189], "provide": [27], "premium": [28], "intelligent": [29], "services.": [30], "Due": [31], "portability": [33], "considerations,": [34], "most": [35], "IoT": [36, 60, 88, 121], "devices": [37], "are": [38], "computing-constrained": [39], "which": [40], "cannot": [41], "afford": [42], "frequent": [43], "DNN": [44], "model": [45, 52, 99], "FL.": [48], "Existing": [49], "approaches": [50], "use": [51], "compression": [53], "techniques": [54], "reduce": [56, 139, 171], "computing": [57, 85, 117, 141, 154, 176, 191], "cost": [58, 118, 155, 177, 192], "whereas": [62], "accuracy": [63, 183], "degradation": [64], "is": [65], "inevitably": [66], "incurred.": [67], "To": [68], "address": [69], "this": [70], "challenge,": [71], "we": [72, 91, 144], "propose": [73], "an": [74, 94, 107, 146], "elastic": [75, 96, 149], "collaboration": [78, 110], "framework,": [79], "namely": [80], "EFLCF,": [81, 114], "accommodate": [83], "limited": [84], "resources": [86], "devices.": [89, 122], "Specifically,": [90], "first": [92], "design": [93, 145], "FL-oriented": [95], "with": [100, 106], "multiple-width": [101], "subnets,": [102], "and": [103, 137, 178], "couple": [104], "it": [105], "FL": [108], "device-server": [109, 134], "framework": [111], "form": [113], "thereby": [115], "releasing": [116], "pressure": [119], "We": [123], "then": [124], "develop": [125], "freezing-assisted": [127], "wide-to-narrow": [128], "mechanism": [130, 151], "realize": [132], "efficient": [133], "distributed": [135], "further": [138], "device": [140, 175], "cost.": [142], "Finally,": [143], "entropy-based": [147], "narrow-to-wide": [148], "inference": [150, 157], "decrease": [153], "without": [158], "compromising": [159], "accuracy.": [160], "Experiments": [161], "demonstrate": [162], "that": [163], "compared": [164], "well-known": [166], "benchmarks,": [167], "our": [168], "EFLCF": [169], "up": [172, 180, 188], "97.65%": [174], "improve": [179], "48.3%": [182], "training,": [185], "while": [186], "reducing": [187], "42.5%": [190], "inference.": [194]}, "primary_topic": {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.9926999807357788, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.9926999807357788, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.005499999970197678, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T13038", "display_name": "Internet of Things and AI", "score": 0.00039999998989515007, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/tcad.2025.3593436", "is_oa": false, "landing_page_url": "https://doi.org/10.1109/tcad.2025.3593436", "pdf_url": null, "source": {"id": "https://openalex.org/S100835903", "display_name": "IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems", "issn_l": "0278-0070", "issn": ["0278-0070", "1937-4151"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-2", "norway-1"], "host_organization": "https://openalex.org/P4310319808", "host_organization_name": "Institute of Electrical and Electronics Engineers", "host_organization_lineage": ["https://openalex.org/P4310319808"], "host_organization_lineage_names": ["Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems", "raw_type": "journal-article"}, "cited_by_count": 3, "is_retracted": false}, {"id": "https://openalex.org/W4416933662", "doi": "https://doi.org/10.1002/cpe.70456", "title": "Dynamic Load Balancing for Distributed Large Model Training: A Hybrid Framework of Gray Markov Chain and MDP", "publication_year": 2025, "publication_date": "2025-12-02", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5100364659", "display_name": "Yonggang Li", "orcid": "https://orcid.org/0009-0005-1836-2450", "observed_orcids": ["https://orcid.org/0009-0005-1836-2450"]}, "institutions": [{"id": "https://openalex.org/I10535382", "display_name": "Chongqing University of Posts and Telecommunications", "ror": "https://ror.org/03dgaqz26", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I10535382"]}], "countries": ["CN"], "is_corresponding": true, "raw_author_name": "Yonggang Li", "raw_affiliation_strings": ["School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China"], "raw_orcid": "https://orcid.org/0009-0005-1836-2450", "affiliations": [{"raw_affiliation_string": "School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China", "institution_ids": ["https://openalex.org/I10535382"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Rui Ji", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I10535382", "display_name": "Chongqing University of Posts and Telecommunications", "ror": "https://ror.org/03dgaqz26", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I10535382"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Rui Ji", "raw_affiliation_strings": ["School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China", "institution_ids": ["https://openalex.org/I10535382"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5101268951", "display_name": "Yaotong Su", "orcid": "https://orcid.org/0009-0006-1271-1426", "observed_orcids": ["https://orcid.org/0009-0006-1271-1426"]}, "institutions": [{"id": "https://openalex.org/I10535382", "display_name": "Chongqing University of Posts and Telecommunications", "ror": "https://ror.org/03dgaqz26", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I10535382"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yaotong Su", "raw_affiliation_strings": ["School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China", "institution_ids": ["https://openalex.org/I10535382"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5115593972", "display_name": "Yuanjin Zhang", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I10535382", "display_name": "Chongqing University of Posts and Telecommunications", "ror": "https://ror.org/03dgaqz26", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I10535382"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yuanjin Zhang", "raw_affiliation_strings": ["School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China", "institution_ids": ["https://openalex.org/I10535382"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5061523006", "display_name": "Andong Zhang", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I10535382", "display_name": "Chongqing University of Posts and Telecommunications", "ror": "https://ror.org/03dgaqz26", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I10535382"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Andong Zhang", "raw_affiliation_strings": ["School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China", "institution_ids": ["https://openalex.org/I10535382"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5103281102", "display_name": "Longjiang Li", "orcid": "https://orcid.org/0000-0003-1388-3300", "observed_orcids": ["https://orcid.org/0000-0003-1388-3300"]}, "institutions": [{"id": "https://openalex.org/I150229711", "display_name": "University of Electronic Science and Technology of China", "ror": "https://ror.org/04qr3zq92", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I150229711"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Longjiang Li", "raw_affiliation_strings": ["School of Information and Communication Engineering University of Electronic Science and Technology  Chengdu China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Information and Communication Engineering University of Electronic Science and Technology  Chengdu China", "institution_ids": ["https://openalex.org/I150229711"]}]}], "abstract_inverted_index": {"ABSTRACT": [0], "Large\u2010scale": [1], "model": [2, 67, 97, 113], "training": [3, 170, 198], "in": [4, 12, 166, 172, 200], "distributed": [5, 169, 197], "data": [6, 65, 115, 134], "centers": [7], "plays": [8], "a": [9, 75], "crucial": [10], "role": [11], "deep": [13, 182], "learning.": [14], "Still,": [15], "it": [16], "faces": [17], "significant": [18], "challenges,": [19, 56], "including": [20], "resource": [21, 131, 147], "fragmentation,": [22], "low": [23], "bandwidth": [24], "utilization,": [25, 148], "and": [26, 50, 69, 80, 91, 133, 149, 179, 194], "complex": [27], "task": [28], "flow": [29, 135], "management.": [30, 122], "The": [31, 95, 123, 156], "problem": [32], "is": [33, 98], "exacerbated": [34], "by": [35], "high\u2010speed,": [36], "high\u2010capacity": [37], "parameter": [38], "synchronization,": [39], "often": [40], "exceeding": [41], "several": [42], "hundred": [43], "Gbps,": [44], "which": [45], "leads": [46], "to": [47, 100, 118, 153], "reduced": [48], "throughput": [49], "computational": [51, 93, 150], "inefficiencies.": [52], "To": [53], "address": [54], "these": [55, 126], "this": [57, 141, 160], "paper": [58], "proposes": [59], "an": [60], "innovative": [61], "approach": [62, 87, 143, 162], "that": [63, 140, 159], "combines": [64], "parallelism,": [66, 68], "dynamic": [70], "load": [71], "balancing.": [72], "By": [73], "integrating": [74], "Gray": [76], "Markov": [77, 81], "Chain": [78], "(GMC)": [79], "Decision": [82], "Process": [83], "(MDP)": [84], "model,": [85], "the": [86, 111, 177, 192], "dynamically": [88], "schedules": [89], "resources": [90], "balances": [92], "loads.": [94], "GMC": [96], "used": [99], "predict": [101], "future": [102], "node": [103], "loads,": [104], "facilitating": [105], "optimal": [106], "weight": [107], "matrix": [108], "decomposition,": [109], "while": [110], "MDP": [112], "adjusts": [114], "transmission": [116], "paths": [117], "optimize": [119], "network": [120], "traffic": [121], "combination": [124], "of": [125, 181, 196], "two": [127], "models": [128], "enhances": [129], "both": [130], "allocation": [132], "optimization.": [136], "Experimental": [137], "results": [138], "demonstrate": [139], "integrated": [142], "significantly": [144, 175], "improves": [145], "throughput,": [146], "efficiency": [151, 193], "compared": [152], "traditional": [154], "methods.": [155], "findings": [157], "suggest": [158], "hybrid": [161], "performs": [163], "exceptionally": [164], "well": [165], "optimizing": [167], "large\u2010scale": [168], "tasks": [171], "multidata\u2010center": [173], "environments,": [174], "improving": [176], "scalability": [178], "performance": [180], "learning": [183], "workloads.": [184], "This": [185], "research": [186], "shows": [187], "promising": [188], "implications": [189], "for": [190], "enhancing": [191], "effectiveness": [195], "systems": [199], "high\u2010demand": [201], "applications.": [202]}, "primary_topic": {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.6875, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.6875, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10715", "display_name": "Distributed and Parallel Computing Systems", "score": 0.20720000565052032, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.07199999690055847, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1002/cpe.70456", "is_oa": false, "landing_page_url": "https://doi.org/10.1002/cpe.70456", "pdf_url": null, "source": {"id": "https://openalex.org/S11065456", "display_name": "Concurrency and Computation Practice and Experience", "issn_l": "1532-0626", "issn": ["1532-0626", "1532-0634"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-1", "norway-1"], "host_organization": "https://openalex.org/P4310320595", "host_organization_name": "Wiley", "host_organization_lineage": ["https://openalex.org/P4310320595"], "host_organization_lineage_names": ["Wiley"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Concurrency and Computation: Practice and Experience", "raw_type": "journal-article"}, "cited_by_count": 1, "is_retracted": false}, {"id": "https://openalex.org/W4411864119", "doi": "https://doi.org/10.1016/j.comnet.2025.111489", "title": "nnWeb: Towards efficient WebGPU-based DNN inference via automatic collaborative offloading", "publication_year": 2025, "publication_date": "2025-07-01", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Jiawei Liu", "orcid": "https://orcid.org/0009-0003-7482-3085", "observed_orcids": ["https://orcid.org/0009-0003-7482-3085"]}, "institutions": [{"id": "https://openalex.org/I76569877", "display_name": "Southeast University", "ror": "https://ror.org/04ct4d772", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76569877"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Jiawei Liu", "raw_affiliation_strings": ["School of Computer Science and Engineering, Southeast University, Nanjing, 211189, China"], "raw_orcid": "https://orcid.org/0009-0003-7482-3085", "affiliations": [{"raw_affiliation_string": "School of Computer Science and Engineering, Southeast University, Nanjing, 211189, China", "institution_ids": ["https://openalex.org/I76569877"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Bing Dong", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I76569877", "display_name": "Southeast University", "ror": "https://ror.org/04ct4d772", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76569877"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Bing Dong", "raw_affiliation_strings": ["School of Computer Science and Engineering, Southeast University, Nanjing, 211189, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Computer Science and Engineering, Southeast University, Nanjing, 211189, China", "institution_ids": ["https://openalex.org/I76569877"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100660988", "display_name": "Weilong Wang", "orcid": "https://orcid.org/0000-0001-6802-1498", "observed_orcids": ["https://orcid.org/0000-0001-6802-1498"]}, "institutions": [{"id": "https://openalex.org/I76569877", "display_name": "Southeast University", "ror": "https://ror.org/04ct4d772", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76569877"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Weilong Wang", "raw_affiliation_strings": ["School of Computer Science and Engineering, Southeast University, Nanjing, 211189, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Computer Science and Engineering, Southeast University, Nanjing, 211189, China", "institution_ids": ["https://openalex.org/I76569877"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Muhan Yuan", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I76569877", "display_name": "Southeast University", "ror": "https://ror.org/04ct4d772", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76569877"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Muhan Yuan", "raw_affiliation_strings": ["School of Computer Science and Engineering, Southeast University, Nanjing, 211189, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Computer Science and Engineering, Southeast University, Nanjing, 211189, China", "institution_ids": ["https://openalex.org/I76569877"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5017989132", "display_name": "Borui Li", "orcid": "https://orcid.org/0000-0001-5262-2483", "observed_orcids": ["https://orcid.org/0000-0001-5262-2483"]}, "institutions": [{"id": "https://openalex.org/I76569877", "display_name": "Southeast University", "ror": "https://ror.org/04ct4d772", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76569877"]}], "countries": ["CN"], "is_corresponding": true, "raw_author_name": "Borui Li", "raw_affiliation_strings": ["School of Computer Science and Engineering, Southeast University, Nanjing, 211189, China"], "raw_orcid": "https://orcid.org/0000-0001-5262-2483", "affiliations": [{"raw_affiliation_string": "School of Computer Science and Engineering, Southeast University, Nanjing, 211189, China", "institution_ids": ["https://openalex.org/I76569877"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5075474174", "display_name": "Zhao-Dong Xu", "orcid": "https://orcid.org/0009-0006-2513-500X", "observed_orcids": ["https://orcid.org/0009-0006-2513-500X"]}, "institutions": [{"id": "https://openalex.org/I76569877", "display_name": "Southeast University", "ror": "https://ror.org/04ct4d772", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76569877"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Zhao-Dong Xu", "raw_affiliation_strings": ["School of Civil Engineering, Southeast University, Nanjing, 211189, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Civil Engineering, Southeast University, Nanjing, 211189, China", "institution_ids": ["https://openalex.org/I76569877"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5100328264", "display_name": "Shuai Wang", "orcid": "https://orcid.org/0000-0002-0866-0308", "observed_orcids": ["https://orcid.org/0000-0002-0866-0308"]}, "institutions": [{"id": "https://openalex.org/I76569877", "display_name": "Southeast University", "ror": "https://ror.org/04ct4d772", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76569877"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Shuai Wang", "raw_affiliation_strings": ["School of Computer Science and Engineering, Southeast University, Nanjing, 211189, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Computer Science and Engineering, Southeast University, Nanjing, 211189, China", "institution_ids": ["https://openalex.org/I76569877"]}]}], "abstract_inverted_index": null, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.3361999988555908, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.3361999988555908, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T12799", "display_name": "Mobile and Web Applications", "score": 0.2856000065803528, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.12780000269412994, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1016/j.comnet.2025.111489", "is_oa": false, "landing_page_url": "https://doi.org/10.1016/j.comnet.2025.111489", "pdf_url": null, "source": {"id": "https://openalex.org/S63392143", "display_name": "Computer Networks", "issn_l": "1389-1286", "issn": ["1389-1286", "1872-7069"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-2", "norway-2"], "host_organization": "https://openalex.org/P4310320990", "host_organization_name": "Elsevier BV", "host_organization_lineage": ["https://openalex.org/P4310320990"], "host_organization_lineage_names": ["Elsevier BV"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Computer Networks", "raw_type": "journal-article"}, "cited_by_count": 2, "is_retracted": false}, {"id": "https://openalex.org/W4410356774", "doi": "https://doi.org/10.2139/ssrn.5254012", "title": "Distributed Large Language Models and Multimodal Learning", "publication_year": 2025, "publication_date": "2025-01-01", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5116857926", "display_name": "Klaus Elli", "orcid": "https://orcid.org/0009-0002-5673-6116", "observed_orcids": ["https://orcid.org/0009-0002-5673-6116"]}, "institutions": [{"id": "https://openalex.org/I100066346", "display_name": "University of Stuttgart", "ror": "https://ror.org/04vnq7t77", "country_code": "DE", "type": "education", "lineage": ["https://openalex.org/I100066346"]}], "countries": ["DE"], "is_corresponding": true, "raw_author_name": "Klaus Elli", "raw_affiliation_strings": ["University of Stuttgart", "University of Stuttgart, Keplerstra\u00dfe 17, D-70174 Stuttgart, Germany"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "University of Stuttgart", "institution_ids": ["https://openalex.org/I100066346"]}, {"raw_affiliation_string": "University of Stuttgart, Keplerstra\u00dfe 17, D-70174 Stuttgart, Germany", "institution_ids": ["https://openalex.org/I100066346"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Sophia Ramirez", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I62916508", "display_name": "Technical University of Munich", "ror": "https://ror.org/02kkvpp62", "country_code": "DE", "type": "education", "lineage": ["https://openalex.org/I62916508"]}], "countries": ["DE"], "is_corresponding": false, "raw_author_name": "Sophia Ramirez", "raw_affiliation_strings": ["Technical University of Munich"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Technical University of Munich", "institution_ids": ["https://openalex.org/I62916508"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5111121361", "display_name": "Arjun Patel", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I223822909", "display_name": "Heidelberg University", "ror": "https://ror.org/038t36y30", "country_code": "DE", "type": "education", "lineage": ["https://openalex.org/I223822909"]}], "countries": ["DE"], "is_corresponding": false, "raw_author_name": "Arjun Patel", "raw_affiliation_strings": ["Heidelberg University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Heidelberg University", "institution_ids": ["https://openalex.org/I223822909"]}]}], "abstract_inverted_index": null, "primary_topic": {"id": "https://openalex.org/T11714", "display_name": "Multimodal Machine Learning Applications", "score": 0.9496999979019165, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T11714", "display_name": "Multimodal Machine Learning Applications", "score": 0.9496999979019165, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.0340999998152256, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.005900000222027302, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.2139/ssrn.5254012", "is_oa": true, "landing_page_url": "https://doi.org/10.2139/ssrn.5254012", "pdf_url": null, "source": {"id": "https://openalex.org/S4210172589", "display_name": "SSRN Electronic Journal", "issn_l": "1556-5068", "issn": ["1556-5068"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I1318003438", "host_organization_name": "RELX Group (Netherlands)", "host_organization_lineage": ["https://openalex.org/I1318003438"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "acceptedVersion", "is_accepted": true, "is_published": false, "raw_source_name": null, "raw_type": "posted-content"}, "cited_by_count": 1, "is_retracted": false}, {"id": "https://openalex.org/W4416344158", "doi": "https://doi.org/10.48550/arxiv.2511.11373", "title": "MarsRL: Advancing Multi-Agent Reasoning System via Reinforcement Learning with Agentic Pipeline Parallelism", "publication_year": 2025, "publication_date": "2025-11-14", "type": "preprint", "language": null, "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5100687258", "display_name": "Shulin Liu", "orcid": "https://orcid.org/0000-0001-5817-8483", "observed_orcids": ["https://orcid.org/0000-0001-5817-8483"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Liu, Shulin", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": null, "display_name": "Du, Dong", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Du, Dong", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5101827453", "display_name": "Tao Yang", "orcid": "https://orcid.org/0000-0003-0459-4195", "observed_orcids": ["https://orcid.org/0000-0003-0459-4195", "https://orcid.org/0000-0002-0714-2904"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Yang, Tao", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": null, "display_name": "Li, Yang", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Li, Yang", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5086341816", "display_name": "Boyu Qiu", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Qiu, Boyu", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Recent": [0], "progress": [1], "in": [2, 33, 60, 102, 121], "large": [3], "language": [4], "models": [5, 62, 73], "(LLMs)": [6], "has": [7], "been": [8], "propelled": [9], "by": [10, 45], "reinforcement": [11, 89], "learning": [12, 90], "with": [13, 92], "verifiable": [14], "rewards": [15], "(RLVR)": [16], "and": [17, 52, 78, 114, 136, 157], "test-time": [18], "scaling.": [19], "However,": [20], "the": [21, 28, 103, 148], "limited": [22], "output": [23], "length": [24], "of": [25, 30, 150], "LLMs": [26], "constrains": [27], "depth": [29], "reasoning": [31, 39, 155, 163], "attainable": [32], "a": [34, 42, 87], "single": [35], "inference": [36], "process.": [37], "Multi-agent": [38], "systems": [40, 156], "offer": [41], "promising": [43], "alternative": [44], "employing": [46], "multiple": [47], "agents": [48, 101], "including": [49], "Solver,": [50], "Verifier,": [51], "Corrector,": [53], "to": [54, 69, 71, 75, 97, 110, 118, 126, 134, 140, 152], "iteratively": [55], "refine": [56], "solutions.": [57], "While": [58], "effective": [59], "closed-source": [61], "like": [63], "Gemini": [64], "2.5": [65], "Pro,": [66], "they": [67], "struggle": [68], "generalize": [70], "open-source": [72], "due": [74], "insufficient": [76], "critic": [77], "correction": [79], "capabilities.": [80], "To": [81], "address": [82], "this,": [83], "we": [84], "propose": [85], "MarsRL,": [86], "novel": [88], "framework": [91], "agentic": [93], "pipeline": [94], "parallelism,": [95], "designed": [96], "jointly": [98], "optimize": [99], "all": [100], "system.": [104], "MarsRL": [105, 128, 151], "introduces": [106], "agent-specific": [107], "reward": [108, 112], "mechanisms": [109], "mitigate": [111], "noise": [113], "employs": [115], "pipeline-inspired": [116], "training": [117], "enhance": [119], "efficiency": [120], "handling": [122], "long": [123], "trajectories.": [124], "Applied": [125], "Qwen3-30B-A3B-Thinking-2507,": [127], "improves": [129], "AIME2025": [130], "accuracy": [131], "from": [132, 138], "86.5%": [133], "93.3%": [135], "BeyondAIME": [137], "64.9%": [139], "73.8%,": [141], "even": [142], "surpassing": [143], "Qwen3-235B-A22B-Thinking-2507.": [144], "These": [145], "findings": [146], "highlight": [147], "potential": [149], "advance": [153], "multi-agent": [154], "broaden": [158], "their": [159], "applicability": [160], "across": [161], "diverse": [162], "tasks.": [164]}, "primary_topic": {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.33980000019073486, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.33980000019073486, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10181", "display_name": "Natural Language Processing Techniques", "score": 0.1517000049352646, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10462", "display_name": "Reinforcement Learning in Robotics", "score": 0.10350000113248825, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2511.11373", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2511.11373", "pdf_url": "https://arxiv.org/pdf/2511.11373", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4406177792", "doi": "https://doi.org/10.48550/arxiv.2501.03306", "title": "The Robustness of Spiking Neural Networks in Federated Learning with Compression Against Non-omniscient Byzantine Attacks", "publication_year": 2025, "publication_date": "2025-01-06", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5062394655", "display_name": "Manh V. Nguyen", "orcid": "https://orcid.org/0009-0003-6909-9766", "observed_orcids": ["https://orcid.org/0009-0003-6909-9766"]}, "institutions": [{"id": "https://openalex.org/I172980758", "display_name": "Kennesaw State University", "ror": "https://ror.org/00jeqjx33", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I172980758"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Nguyen, Manh V.", "raw_affiliation_strings": ["College of Computing and Software Engineering , Kennesaw State University , Georgia , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "College of Computing and Software Engineering , Kennesaw State University , Georgia , USA", "institution_ids": ["https://openalex.org/I172980758"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5012329541", "display_name": "Liang Zhao", "orcid": "https://orcid.org/0000-0003-0678-489X", "observed_orcids": ["https://orcid.org/0000-0003-0678-489X"]}, "institutions": [{"id": "https://openalex.org/I172980758", "display_name": "Kennesaw State University", "ror": "https://ror.org/00jeqjx33", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I172980758"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Zhao, Liang", "raw_affiliation_strings": ["College of Computing and Software Engineering , Kennesaw State University , Georgia , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "College of Computing and Software Engineering , Kennesaw State University , Georgia , USA", "institution_ids": ["https://openalex.org/I172980758"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100940541", "display_name": "Bobin Deng", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I172980758", "display_name": "Kennesaw State University", "ror": "https://ror.org/00jeqjx33", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I172980758"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Deng, Bobin", "raw_affiliation_strings": ["College of Computing and Software Engineering , Kennesaw State University , Georgia , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "College of Computing and Software Engineering , Kennesaw State University , Georgia , USA", "institution_ids": ["https://openalex.org/I172980758"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5100743605", "display_name": "Shaoen Wu", "orcid": "https://orcid.org/0000-0002-4768-6930", "observed_orcids": ["https://orcid.org/0000-0002-4768-6930"]}, "institutions": [{"id": "https://openalex.org/I172980758", "display_name": "Kennesaw State University", "ror": "https://ror.org/00jeqjx33", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I172980758"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Wu, Shaoen", "raw_affiliation_strings": ["College of Computing and Software Engineering , Kennesaw State University , Georgia , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "College of Computing and Software Engineering , Kennesaw State University , Georgia , USA", "institution_ids": ["https://openalex.org/I172980758"]}]}], "abstract_inverted_index": {"Spiking": [0], "Neural": [1], "Networks": [2], "(SNNs),": [3], "which": [4, 15], "offer": [5], "exceptional": [6], "energy": [7], "efficiency": [8], "for": [9, 66], "inference,": [10], "and": [11, 42, 54, 94, 129], "Federated": [12], "Learning": [13], "(FL),": [14], "offers": [16], "privacy-preserving": [17], "distributed": [18], "training,": [19], "is": [20], "a": [21, 64, 104, 145], "rising": [22], "area": [23], "of": [24, 31, 68, 79, 107, 119, 134, 148], "interest": [25], "that": [26, 38, 86, 103], "highly": [27], "beneficial": [28], "towards": [29], "Internet": [30], "Things": [32], "(IoT)": [33], "devices.": [34], "Despite": [35], "this,": [36], "research": [37], "tackles": [39], "Byzantine": [40, 83, 139], "attacks": [41], "bandwidth": [43, 127], "limitation": [44], "in": [45, 71, 123, 153], "FL-SNNs,": [46, 80], "both": [47, 67, 124], "poses": [48], "significant": [49], "threats": [50], "on": [51], "model": [52], "convergence": [53], "training": [55, 136, 155], "times,": [56], "still": [57], "remains": [58], "largely": [59], "unexplored.": [60], "Going": [61], "beyond": [62], "proposing": [63], "solution": [65], "these": [69], "problems,": [70], "this": [72], "work": [73], "we": [74, 101, 143], "highlight": [75], "the": [76, 111, 117, 120, 132, 157], "dual": [77], "benefits": [78], "against": [81, 137], "non-omniscient": [82, 138], "adversaries": [84], "(ones": [85], "restrict": [87], "attackers": [88], "access": [89], "to": [90], "local": [91], "clients": [92], "datasets),": [93], "greater": [95], "communication": [96], "efficiency,": [97], "over": [98], "FL-ANNs.": [99], "Specifically,": [100], "discovered": [102], "simple": [105], "integration": [106], "Top-\\k{appa}": [108], "sparsification": [109], "into": [110], "FL": [112, 135], "apparatus": [113], "can": [114], "help": [115], "leverage": [116], "advantages": [118], "SNN": [121], "models": [122], "greatly": [125], "reducing": [126], "usage": [128], "significantly": [130], "boosting": [131], "robustness": [133], "adversaries.": [140], "Most": [141], "notably,": [142], "saw": [144], "massive": [146], "improvement": [147], "roughly": [149], "40%": [150], "accuracy": [151], "gain": [152], "FL-SNNs": [154], "under": [156], "lethal": [158], "MinMax": [159], "attack": [160]}, "primary_topic": {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.9970999956130981, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.9970999956130981, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11689", "display_name": "Adversarial Robustness in Machine Learning", "score": 0.00039999998989515007, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.00019999999494757503, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2501.03306", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2501.03306", "pdf_url": "https://arxiv.org/pdf/2501.03306", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4410949929", "doi": "https://doi.org/10.1109/tetc.2025.3573522", "title": "PipeDAP: An Efficient Communication Framework for Scheduling Decoupled All-Reduce Primitives in Distributed DNN Training", "publication_year": 2025, "publication_date": "2025-06-02", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5011127712", "display_name": "Yunqi Gao", "orcid": "https://orcid.org/0009-0003-0599-8383", "observed_orcids": ["https://orcid.org/0009-0003-0599-8383", "https://orcid.org/0009-0004-4194-665X"]}, "institutions": [{"id": "https://openalex.org/I76130692", "display_name": "Zhejiang University", "ror": "https://ror.org/00a2xv884", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76130692"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yunqi Gao", "raw_affiliation_strings": ["Zhejiang University, Hangzhou, China"], "raw_orcid": "https://orcid.org/0009-0003-0599-8383", "affiliations": [{"raw_affiliation_string": "Zhejiang University, Hangzhou, China", "institution_ids": ["https://openalex.org/I76130692"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5102834373", "display_name": "Bing Hu", "orcid": "https://orcid.org/0000-0002-0594-8998", "observed_orcids": ["https://orcid.org/0000-0002-0594-8998"]}, "institutions": [{"id": "https://openalex.org/I76130692", "display_name": "Zhejiang University", "ror": "https://ror.org/00a2xv884", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76130692"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Bing Hu", "raw_affiliation_strings": ["Zhejiang University, Hangzhou, China"], "raw_orcid": "https://orcid.org/0000-0002-0594-8998", "affiliations": [{"raw_affiliation_string": "Zhejiang University, Hangzhou, China", "institution_ids": ["https://openalex.org/I76130692"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5028862760", "display_name": "Mahdi Boloursaz Mashhadi", "orcid": "https://orcid.org/0000-0001-9948-9165", "observed_orcids": ["https://orcid.org/0000-0001-9948-9165"]}, "institutions": [{"id": "https://openalex.org/I28290843", "display_name": "University of Surrey", "ror": "https://ror.org/00ks66431", "country_code": "GB", "type": "education", "lineage": ["https://openalex.org/I28290843"]}], "countries": ["GB"], "is_corresponding": false, "raw_author_name": "Mahdi Boloursaz Mashhadi", "raw_affiliation_strings": ["5GIC &#x0026; 6GIC, Institute for Communication Systems (ICS), University of Surrey, Guildford, U.K"], "raw_orcid": "https://orcid.org/0000-0001-9948-9165", "affiliations": [{"raw_affiliation_string": "5GIC &#x0026; 6GIC, Institute for Communication Systems (ICS), University of Surrey, Guildford, U.K", "institution_ids": ["https://openalex.org/I28290843"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100392219", "display_name": "Wei Wang", "orcid": "https://orcid.org/0000-0003-2153-9075", "observed_orcids": ["https://orcid.org/0000-0003-2153-9075", "https://orcid.org/0000-0001-9032-4401"]}, "institutions": [{"id": "https://openalex.org/I76130692", "display_name": "Zhejiang University", "ror": "https://ror.org/00a2xv884", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I76130692"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Wei Wang", "raw_affiliation_strings": ["Zhejiang University, Hangzhou, China"], "raw_orcid": "https://orcid.org/0000-0003-2153-9075", "affiliations": [{"raw_affiliation_string": "Zhejiang University, Hangzhou, China", "institution_ids": ["https://openalex.org/I76130692"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5032549075", "display_name": "Rahim Tafazolli", "orcid": "https://orcid.org/0000-0002-6062-8639", "observed_orcids": ["https://orcid.org/0000-0002-6062-8639"]}, "institutions": [{"id": "https://openalex.org/I28290843", "display_name": "University of Surrey", "ror": "https://ror.org/00ks66431", "country_code": "GB", "type": "education", "lineage": ["https://openalex.org/I28290843"]}], "countries": ["GB"], "is_corresponding": false, "raw_author_name": "Rahim Tafazolli", "raw_affiliation_strings": ["5GIC &#x0026; 6GIC, Institute for Communication Systems (ICS), University of Surrey, Guildford, U.K"], "raw_orcid": "https://orcid.org/0000-0002-6062-8639", "affiliations": [{"raw_affiliation_string": "5GIC &#x0026; 6GIC, Institute for Communication Systems (ICS), University of Surrey, Guildford, U.K", "institution_ids": ["https://openalex.org/I28290843"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5056145687", "display_name": "Merouane Abdelkader Debbah", "orcid": "https://orcid.org/0000-0001-8941-8080", "observed_orcids": ["https://orcid.org/0000-0001-8941-8080"]}, "institutions": [{"id": "https://openalex.org/I176601375", "display_name": "Khalifa University of Science and Technology", "ror": "https://ror.org/05hffr360", "country_code": "AE", "type": "education", "lineage": ["https://openalex.org/I176601375"]}], "countries": ["AE"], "is_corresponding": false, "raw_author_name": "M\u00e9rouane Debbah", "raw_affiliation_strings": ["KU 6G Research Center, Department of Computer and Information Engineering, Khalifa University, Abu Dhabi, UAE"], "raw_orcid": "https://orcid.org/0000-0001-8941-8080", "affiliations": [{"raw_affiliation_string": "KU 6G Research Center, Department of Computer and Information Engineering, Khalifa University, Abu Dhabi, UAE", "institution_ids": ["https://openalex.org/I176601375"]}]}], "abstract_inverted_index": {"Communication": [0], "scheduling": [1, 21, 62, 110, 179], "effectively": [2], "improves": [3], "the": [4, 36, 54, 90, 101, 108, 113, 131, 136, 176], "scalability": [5], "of": [6, 95, 112, 123, 148, 171], "distributed": [7, 146], "deep": [8], "learning": [9], "by": [10], "overlapping": [11], "computation": [12], "and": [13, 49, 92, 106, 115, 125, 142, 166], "communication": [14, 20, 55, 75, 122, 178], "tasks": [15], "during": [16], "training.": [17], "However,": [18], "existing": [19], "frameworks": [22], "based": [23], "on": [24, 139, 154], "tensor": [25], "partitioning": [26, 33, 51], "suffer": [27], "from": [28], "two": [29, 155], "fundamental": [30], "issues:": [31], "(1)": [32], "schemes": [34, 52], "at": [35, 53], "data": [37], "volume": [38], "level": [39, 57], "introduce": [40], "extensive": [41], "startup": [42, 132], "overheads": [43], "leading": [44], "to": [45, 88, 128, 163, 169, 175], "higher": [46], "energy": [47, 93, 172], "consumption,": [48], "(2)": [50], "primitive": [56], "do": [58], "not": [59], "provide": [60], "optimal": [61], "resulting": [63], "in": [64, 84], "longer": [65], "training": [66, 96, 147], "time.": [67], "In": [68], "this": [69], "paper,": [70], "we": [71, 119], "propose": [72], "an": [73], "efficient": [74], "mechanism,": [76], "namely": [77], "PipeDAP,": [78], "which": [79], "schedules": [80], "decoupled": [81], "all-reduce": [82], "operations": [83, 127], "a": [85], "near-optimal": [86, 109], "order": [87, 111], "minimize": [89], "time": [91], "consumption": [94, 173], "DNN": [97, 150], "models.": [98, 151], "We": [99, 134], "build": [100], "mathematical": [102], "model": [103], "for": [104, 145], "PipeDAP": [105, 137, 160], "derive": [107], "reduce-scatter": [114, 124], "all-gather": [116, 126], "operations.": [117], "Meanwhile,": [118], "leverage": [120], "simultaneous": [121], "further": [129], "reduce": [130], "overheads.": [133], "implement": [135], "architecture": [138], "PyTorch": [140], "framework,": [141], "apply": [143], "it": [144], "benchmark": [149], "Experimental": [152], "results": [153], "GPU": [156], "clusters": [157], "demonstrate": [158], "that": [159], "achieves": [161], "up": [162, 168], "1.82x": [164], "speedup": [165], "saves": [167], "45.4%": [170], "compared": [174], "state-of-the-art": [177], "frameworks.": [180], "PipeDAP's": [181], "code": [182], "is": [183], "available": [184], "athttps://github.com/ZJU-CNLAB/PipeDAP.": [185]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9613999724388123, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9613999724388123, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.006000000052154064, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10715", "display_name": "Distributed and Parallel Computing Systems", "score": 0.00570000009611249, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/tetc.2025.3573522", "is_oa": false, "landing_page_url": "https://doi.org/10.1109/tetc.2025.3573522", "pdf_url": null, "source": {"id": "https://openalex.org/S2496326734", "display_name": "IEEE Transactions on Emerging Topics in Computing", "issn_l": "2168-6750", "issn": ["2168-6750", "2376-4562"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-1", "norway-1"], "host_organization": "https://openalex.org/P4310319808", "host_organization_name": "Institute of Electrical and Electronics Engineers", "host_organization_lineage": ["https://openalex.org/P4310319808"], "host_organization_lineage_names": ["Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Transactions on Emerging Topics in Computing", "raw_type": "journal-article"}, "cited_by_count": 1, "is_retracted": false}, {"id": "https://openalex.org/W4415130124", "doi": "https://doi.org/10.48550/arxiv.2506.02007", "title": "eACGM: Non-instrumented Performance Tracing and Anomaly Detection towards Machine Learning Systems", "publication_year": 2025, "publication_date": "2025-05-25", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5100958778", "display_name": "Ruilin Xu", "orcid": "https://orcid.org/0009-0009-0141-2545", "observed_orcids": ["https://orcid.org/0009-0009-0141-2545", "https://orcid.org/0009-0006-9320-8012"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Xu, Ruilin", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": null, "display_name": "Xie, Zongxuan", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Xie, Zongxuan", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5023707916", "display_name": "Pengfei Chen", "orcid": "https://orcid.org/0000-0001-8306-0081", "observed_orcids": ["https://orcid.org/0000-0001-8306-0081", "https://orcid.org/0009-0000-7319-3868"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Chen, Pengfei", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"We": [0], "present": [1], "eACGM,": [2, 126], "a": [3, 62, 129], "full-stack": [4], "AI/ML": [5, 173], "system": [6, 98], "monitoring": [7, 151], "framework": [8], "based": [9], "on": [10], "eBPF.": [11], "eACGM": [12, 79], "collects": [13], "real-time": [14], "performance": [15, 71, 137, 148, 166], "data": [16], "from": [17, 31], "key": [18, 32], "hardware": [19, 89], "components,": [20], "including": [21], "the": [22, 68], "GPU": [23, 56], "and": [24, 39, 76, 91, 100, 107, 114, 131, 142, 149, 156, 168], "network": [25], "communication": [26, 92], "layer,": [27], "as": [28, 30, 36, 86], "well": [29], "software": [33], "stacks": [34], "such": [35, 85], "CUDA,": [37], "Python,": [38], "PyTorch,": [40], "all": [41], "without": [42], "requiring": [43], "any": [44], "code": [45], "instrumentation": [46], "or": [47], "modifications.": [48], "Additionally,": [49], "it": [50], "leverages": [51], "libnvml": [52], "to": [53, 67], "gather": [54], "process-level": [55], "resource": [57], "usage": [58], "information.": [59], "By": [60], "applying": [61], "Gaussian": [63], "Mixture": [64], "Model": [65], "(GMM)": [66], "collected": [69], "multidimensional": [70], "metrics": [72], "for": [73, 165], "statistical": [74], "modeling": [75], "clustering": [77], "analysis,": [78], "effectively": [80], "identifies": [81], "complex": [82], "failure": [83], "modes,": [84], "latency": [87], "anomalies,": [88], "failures,": [90], "inefficiencies,": [93], "enabling": [94], "rapid": [95], "diagnosis": [96, 170], "of": [97], "bottlenecks": [99], "abnormal": [101], "behaviors.": [102], "To": [103], "evaluate": [104], "eACGM's": [105], "effectiveness": [106], "practicality,": [108], "we": [109], "conducted": [110], "extensive": [111], "empirical": [112], "studies": [113], "case": [115], "analyses": [116], "in": [117, 158, 171], "multi-node": [118], "distributed": [119], "training": [120, 141], "scenarios.": [121], "The": [122], "results": [123], "demonstrate": [124], "that": [125], "while": [127], "maintaining": [128], "non-intrusive": [130], "low-overhead": [132], "profile,": [133], "successfully": [134], "captures": [135], "critical": [136], "anomalies": [138], "during": [139], "model": [140], "inference.": [143], "Its": [144], "stable": [145], "anomaly": [146], "detection": [147], "comprehensive": [150], "capabilities": [152], "validate": [153], "its": [154], "applicability": [155], "scalability": [157], "real-world": [159], "production": [160], "environments,": [161], "providing": [162], "strong": [163], "support": [164], "optimization": [167], "fault": [169], "large-scale": [172], "systems.": [174]}, "primary_topic": {"id": "https://openalex.org/T12127", "display_name": "Software System Performance and Reliability", "score": 0.9976000189781189, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T12127", "display_name": "Software System Performance and Reliability", "score": 0.9976000189781189, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.0019000000320374966, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 9.999999747378752e-05, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2506.02007", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2506.02007", "pdf_url": "https://arxiv.org/pdf/2506.02007", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4408780788", "doi": "https://doi.org/10.36227/techrxiv.174284672.24268959/v1", "title": "ESL-LEO: An Efficient Split Learning Framework over LEO Satellite Networks", "publication_year": 2025, "publication_date": "2025-03-24", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5004434493", "display_name": "Zheng Lin", "orcid": "https://orcid.org/0009-0006-4977-1385", "observed_orcids": ["https://orcid.org/0009-0006-4977-1385"]}, "institutions": [{"id": "https://openalex.org/I24943067", "display_name": "Fudan University", "ror": "https://ror.org/013q1eq08", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I24943067"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Zheng Lin", "raw_affiliation_strings": ["Institute of Space Internet, Fudan University", "School of Computer Science, Fudan University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Institute of Space Internet, Fudan University", "institution_ids": ["https://openalex.org/I24943067"]}, {"raw_affiliation_string": "School of Computer Science, Fudan University", "institution_ids": ["https://openalex.org/I24943067"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5056773892", "display_name": "Y. Zhang", "orcid": "https://orcid.org/0000-0001-7349-3780", "observed_orcids": ["https://orcid.org/0000-0001-7349-3780"]}, "institutions": [{"id": "https://openalex.org/I24943067", "display_name": "Fudan University", "ror": "https://ror.org/013q1eq08", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I24943067"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yuxin Zhang", "raw_affiliation_strings": ["Institute of Space Internet, Fudan University", "School of Computer Science, Fudan University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Institute of Space Internet, Fudan University", "institution_ids": ["https://openalex.org/I24943067"]}, {"raw_affiliation_string": "School of Computer Science, Fudan University", "institution_ids": ["https://openalex.org/I24943067"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100656819", "display_name": "Zhe Chen", "orcid": "https://orcid.org/0000-0002-3215-2696", "observed_orcids": ["https://orcid.org/0000-0002-3215-2696", "https://orcid.org/0000-0003-2286-4988"]}, "institutions": [{"id": "https://openalex.org/I24943067", "display_name": "Fudan University", "ror": "https://ror.org/013q1eq08", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I24943067"]}], "countries": ["CN"], "is_corresponding": true, "raw_author_name": "Zhe Chen", "raw_affiliation_strings": ["Institute of Space Internet, Fudan University", "School of Computer Science, Fudan University"], "raw_orcid": "https://orcid.org/0000-0002-3215-2696", "affiliations": [{"raw_affiliation_string": "Institute of Space Internet, Fudan University", "institution_ids": ["https://openalex.org/I24943067"]}, {"raw_affiliation_string": "School of Computer Science, Fudan University", "institution_ids": ["https://openalex.org/I24943067"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5057230946", "display_name": "Zihan Fang", "orcid": "https://orcid.org/0000-0002-0844-9879", "observed_orcids": ["https://orcid.org/0000-0002-0844-9879"]}, "institutions": [{"id": "https://openalex.org/I24943067", "display_name": "Fudan University", "ror": "https://ror.org/013q1eq08", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I24943067"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Zihan Fang", "raw_affiliation_strings": ["Institute of Space Internet, Fudan University", "School of Computer Science, Fudan University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Institute of Space Internet, Fudan University", "institution_ids": ["https://openalex.org/I24943067"]}, {"raw_affiliation_string": "School of Computer Science, Fudan University", "institution_ids": ["https://openalex.org/I24943067"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Zhe Chen", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I154099455", "display_name": "Shandong University", "ror": "https://ror.org/0207yh398", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I154099455"]}, {"id": "https://openalex.org/I24943067", "display_name": "Fudan University", "ror": "https://ror.org/013q1eq08", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I24943067"]}], "countries": ["CN"], "is_corresponding": true, "raw_author_name": "Zhe Chen", "raw_affiliation_strings": ["Institute of Space Internet, Fudan University", "School of Computer Science, Fudan University", "School of Computer Science and Technology, Shandong University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Institute of Space Internet, Fudan University", "institution_ids": ["https://openalex.org/I24943067"]}, {"raw_affiliation_string": "School of Computer Science, Fudan University", "institution_ids": ["https://openalex.org/I24943067"]}, {"raw_affiliation_string": "School of Computer Science and Technology, Shandong University", "institution_ids": ["https://openalex.org/I154099455"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Zihan Fang", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I154099455", "display_name": "Shandong University", "ror": "https://ror.org/0207yh398", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I154099455"]}, {"id": "https://openalex.org/I24943067", "display_name": "Fudan University", "ror": "https://ror.org/013q1eq08", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I24943067"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Zihan Fang", "raw_affiliation_strings": ["Institute of Space Internet, Fudan University", "School of Computer Science, Fudan University", "School of Computer Science and Technology, Shandong University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Institute of Space Internet, Fudan University", "institution_ids": ["https://openalex.org/I24943067"]}, {"raw_affiliation_string": "School of Computer Science, Fudan University", "institution_ids": ["https://openalex.org/I24943067"]}, {"raw_affiliation_string": "School of Computer Science and Technology, Shandong University", "institution_ids": ["https://openalex.org/I154099455"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Yanni Yang", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I108688024", "display_name": "Qingdao University", "ror": "https://ror.org/021cj6z65", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I108688024"]}, {"id": "https://openalex.org/I154099455", "display_name": "Shandong University", "ror": "https://ror.org/0207yh398", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I154099455"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yanni Yang", "raw_affiliation_strings": ["School of Computer Science and Technology, Shandong University", "College of Computer Science and Technology, Qingdao University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Computer Science and Technology, Shandong University", "institution_ids": ["https://openalex.org/I154099455"]}, {"raw_affiliation_string": "College of Computer Science and Technology, Qingdao University", "institution_ids": ["https://openalex.org/I108688024"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100618997", "display_name": "Guoming Zhang", "orcid": "https://orcid.org/0000-0003-2123-8125", "observed_orcids": ["https://orcid.org/0000-0003-2123-8125"]}, "institutions": [{"id": "https://openalex.org/I154099455", "display_name": "Shandong University", "ror": "https://ror.org/0207yh398", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I154099455"]}, {"id": "https://openalex.org/I889458895", "display_name": "University of Hong Kong", "ror": "https://ror.org/02zhqgq86", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I889458895"]}], "countries": ["CN", "HK"], "is_corresponding": false, "raw_author_name": "Guoming Zhang", "raw_affiliation_strings": ["School of Computer Science and Technology, Shandong University", "Department of Electrical and Electronic Engineering, University of Hong Kong"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Computer Science and Technology, Shandong University", "institution_ids": ["https://openalex.org/I154099455"]}, {"raw_affiliation_string": "Department of Electrical and Electronic Engineering, University of Hong Kong", "institution_ids": ["https://openalex.org/I889458895"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Huan Yang", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I108688024", "display_name": "Qingdao University", "ror": "https://ror.org/021cj6z65", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I108688024"]}, {"id": "https://openalex.org/I889458895", "display_name": "University of Hong Kong", "ror": "https://ror.org/02zhqgq86", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I889458895"]}], "countries": ["CN", "HK"], "is_corresponding": false, "raw_author_name": "Huan Yang", "raw_affiliation_strings": ["College of Computer Science and Technology, Qingdao University", "Department of Electrical and Electronic Engineering, University of Hong Kong"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "College of Computer Science and Technology, Qingdao University", "institution_ids": ["https://openalex.org/I108688024"]}, {"raw_affiliation_string": "Department of Electrical and Electronic Engineering, University of Hong Kong", "institution_ids": ["https://openalex.org/I889458895"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5101886946", "display_name": "Cong Wu", "orcid": "https://orcid.org/0000-0003-1360-1617", "observed_orcids": ["https://orcid.org/0000-0003-1360-1617"]}, "institutions": [{"id": "https://openalex.org/I24943067", "display_name": "Fudan University", "ror": "https://ror.org/013q1eq08", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I24943067"]}, {"id": "https://openalex.org/I889458895", "display_name": "University of Hong Kong", "ror": "https://ror.org/02zhqgq86", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I889458895"]}], "countries": ["CN", "HK"], "is_corresponding": false, "raw_author_name": "Cong Wu", "raw_affiliation_strings": ["Department of Electrical and Electronic Engineering, University of Hong Kong", "Institute of Space Internet, Fudan University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Electrical and Electronic Engineering, University of Hong Kong", "institution_ids": ["https://openalex.org/I889458895"]}, {"raw_affiliation_string": "Institute of Space Internet, Fudan University", "institution_ids": ["https://openalex.org/I24943067"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5083484070", "display_name": "Xianhao Chen", "orcid": "https://orcid.org/0000-0002-4295-940X", "observed_orcids": ["https://orcid.org/0000-0002-4295-940X", "https://orcid.org/0000-0001-5401-0157", "https://orcid.org/0009-0001-7031-1440"]}, "institutions": [{"id": "https://openalex.org/I889458895", "display_name": "University of Hong Kong", "ror": "https://ror.org/02zhqgq86", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I889458895"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Xianhao Chen", "raw_affiliation_strings": ["Department of Electrical and Electronic Engineering, University of Hong Kong"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Electrical and Electronic Engineering, University of Hong Kong", "institution_ids": ["https://openalex.org/I889458895"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5075109962", "display_name": "Yue Gao", "orcid": "https://orcid.org/0000-0001-6502-9910", "observed_orcids": ["https://orcid.org/0000-0001-6502-9910", "https://orcid.org/0000-0002-6027-5618"]}, "institutions": [{"id": "https://openalex.org/I24943067", "display_name": "Fudan University", "ror": "https://ror.org/013q1eq08", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I24943067"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yue Gao", "raw_affiliation_strings": ["Institute of Space Internet, Fudan University", "School of Computer Science, Fudan University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Institute of Space Internet, Fudan University", "institution_ids": ["https://openalex.org/I24943067"]}, {"raw_affiliation_string": "School of Computer Science, Fudan University", "institution_ids": ["https://openalex.org/I24943067"]}]}], "abstract_inverted_index": {"Recently,": [0], "a": [1, 85], "large": [2], "number": [3], "of": [4, 28, 143], "low": [5], "Earth": [6], "orbit": [7], "(LEO)": [8], "satellites": [9, 20, 61], "have": [10], "been": [11], "launched": [12], "and": [13, 62, 170, 191], "deployed": [14], "successfully": [15], "in": [16, 107, 188], "space.": [17], "These": [18], "LEO": [19, 124], "are": [21], "equipped": [22], "with": [23], "advanced": [24], "multimodal": [25], "sensors": [26], "capable": [27], "collecting": [29], "massive": [30, 50], "sensor": [31], "data,": [32], "serving": [33], "as": [34, 84], "fuel": [35], "for": [36, 53, 123, 168], "various": [37], "space-based": [38], "deep": [39], "learning": [40, 80], "(DL)": [41], "applications.": [42], "However,": [43], "ground": [44], "stations": [45], "(GS)": [46], "cannot": [47], "download": [48], "such": [49], "raw": [51], "data": [52], "centralized": [54], "training": [55, 74, 91, 141, 178], "due": [56], "to": [57, 72, 88, 93, 127, 138, 152, 164], "intermittent": [58], "connectivity": [59, 102], "between": [60], "GS,": [63], "while": [64], "the": [65, 98, 140, 144, 159, 172], "scaled-up": [66], "DL": [67], "models": [68], "pose": [69], "substantial": [70], "barriers": [71], "distributed": [73], "on": [75, 100], "resource-constrained": [76], "satellites.": [77], "Though": [78], "split": [79], "(SL)": [81], "has": [82], "emerged": [83], "promising": [86], "solution": [87], "offload": [89], "major": [90], "workloads": [92], "GS": [94], "via": [95], "model": [96, 137, 154], "partitioning,": [97], "reliance": [99], "continuous": [101], "significantly": [103], "limits": [104], "its": [105], "deployment": [106], "satellite": [108, 125], "networks.": [109], "To": [110], "address": [111], "this": [112], "challenge,": [113], "we": [114, 157], "propose": [115], "ESL-LEO,": [116], "an": [117, 135], "efficient": [118], "SL": [119], "framework": [120], "specifically": [121], "tailored": [122], "networks": [126], "enable": [128], "full-satellite": [129], "period": [130], "training.": [131], "We": [132], "first": [133], "construct": [134], "auxiliary": [136], "tackle": [139], "failure": [142], "satellite-GS": [145], "non-contact": [146], "time,": [147], "leveraging": [148], "on-satellite": [149], "idle": [150], "resources": [151], "expedite": [153], "convergence.": [155], "Moreover,": [156], "design": [158], "online": [160], "knowledge": [161, 167], "distillation": [162], "mechanism": [163], "extract": [165], "GS-side": [166], "guiding": [169], "refining": [171], "satellite-side": [173], "training,": [174], "thereby": [175], "further": [176], "enhancing": [177], "performance.": [179], "Extensive": [180], "experiments": [181], "demonstrate": [182], "that": [183], "ESL-LEO": [184], "outperforms": [185], "state-of-the-art": [186], "benchmarks": [187], "both": [189], "accuracy": [190], "convergence": [192], "speed.": [193]}, "primary_topic": {"id": "https://openalex.org/T12042", "display_name": "Satellite Communication Systems", "score": 0.9580000042915344, "subfield": {"id": "https://openalex.org/subfields/2202", "display_name": "Aerospace Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T12042", "display_name": "Satellite Communication Systems", "score": 0.9580000042915344, "subfield": {"id": "https://openalex.org/subfields/2202", "display_name": "Aerospace Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.015699999406933784, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11701", "display_name": "Space Satellite Systems and Control", "score": 0.002300000051036477, "subfield": {"id": "https://openalex.org/subfields/2202", "display_name": "Aerospace Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.36227/techrxiv.174284672.24268959/v1", "is_oa": true, "landing_page_url": "https://doi.org/10.36227/techrxiv.174284672.24268959/v1", "pdf_url": null, "source": null, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "acceptedVersion", "is_accepted": true, "is_published": false, "raw_source_name": null, "raw_type": "posted-content"}, "cited_by_count": 3, "is_retracted": false}, {"id": "https://openalex.org/W7110975370", "doi": "https://doi.org/10.12732/ijam.v38i11s.1277", "title": "AI-DRIVEN ANOMALY DETECTION IN IOT NETWORKS USING ADVANCED MACHINE LEARNING TECHNIQUES", "publication_year": 2025, "publication_date": "2025-11-09", "type": "article", "language": null, "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Azyk Orozonova", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I17831865", "display_name": "Kyrgyz State Technical University", "ror": "https://ror.org/01c9f4y89", "country_code": "KG", "type": "education", "lineage": ["https://openalex.org/I17831865"]}], "countries": ["KG"], "is_corresponding": true, "raw_author_name": "Azyk Orozonova", "raw_affiliation_strings": ["Deputy Director for Research , International Higher School of Logistics , Kyrgyz State Technical University named after I. Razzakov , 66 Ch. Aitmatov Avenue , Bishkek 720044 , Kyrgyz Republic"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Deputy Director for Research , International Higher School of Logistics , Kyrgyz State Technical University named after I. Razzakov , 66 Ch. Aitmatov Avenue , Bishkek 720044 , Kyrgyz Republic", "institution_ids": ["https://openalex.org/I17831865"]}]}], "abstract_inverted_index": {"The": [0, 59, 79, 130, 168], "rapid": [1], "expansion": [2], "of": [3, 5, 148, 174], "Internet": [4], "Things": [6], "(IoT)": [7], "ecosystems": [8], "has": [9], "significantly": [10], "increased": [11], "the": [12, 145, 149, 172], "attack": [13], "surface,": [14], "making": [15], "real-time": [16], "anomaly": [17, 34], "detection": [18, 35, 120, 153], "a": [19, 31, 183], "critical": [20], "requirement": [21], "for": [22, 164, 187], "ensuring": [23], "network": [24, 190], "resilience": [25, 126], "and": [26, 44, 52, 90, 96, 104, 115, 125, 141], "security.": [27, 191], "This": [28], "research": [29], "presents": [30], "hybrid": [32, 131], "AI-driven": [33], "architecture": [36], "integrating": [37], "Graph": [38], "Neural": [39], "Networks": [40], "(GNNs),": [41], "Transformer": [42], "encoders,": [43], "Autoencoder-based": [45], "reconstruction": [46], "learning": [47, 103, 107], "to": [48, 100, 118, 127], "capture": [49], "spatial,": [50], "temporal,": [51], "behavioral": [53], "dependencies": [54], "within": [55], "heterogeneous": [56], "IoT": [57, 72, 166, 189], "traffic.": [58], "model": [60, 123, 132, 150], "is": [61, 82], "further": [62], "enhanced": [63], "through": [64], "federated": [65, 116, 146], "learning,": [66], "enabling": [67], "privacy-preserving": [68], "distributed": [69], "training": [70], "across": [71, 113], "devices": [73], "while": [74], "maintaining": [75], "strong": [76], "predictive": [77], "performance.": [78], "proposed": [80], "system": [81], "evaluated": [83], "using": [84], "benchmark": [85], "datasets": [86], "such": [87], "as": [88], "N-BaIoT": [89], "Bot-IoT,": [91], "demonstrating": [92], "superior": [93], "accuracy,": [94], "robustness,": [95], "cross-device": [97], "generalization": [98], "compared": [99], "traditional": [101], "machine": [102], "standalone": [105], "deep": [106], "methods.": [108], "Comprehensive": [109], "experiments": [110], "were": [111], "conducted": [112], "centralized": [114], "environments": [117], "assess": [119], "performance,": [121], "scalability,": [122], "stability,": [124], "concept": [128], "drift.": [129], "consistently": [133], "achieved": [134], "F1-scores": [135], "above": [136], "98%,": [137], "outperforming": [138], "GNN-only,": [139], "Transformer-only,": [140], "Autoencoder-only": [142], "baselines.": [143], "Furthermore,": [144], "version": [147], "preserved": [151], "high": [152], "accuracy": [154], "(99.1%)": [155], "even": [156], "under": [157], "non-IID": [158], "data": [159], "distributions,": [160], "validating": [161], "its": [162], "suitability": [163], "privacy-sensitive": [165], "deployments.": [167], "results": [169], "indicate": [170], "that": [171], "fusion": [173], "multiple": [175], "AI": [176], "techniques,": [177], "combined": [178], "with": [179], "decentralized": [180], "training,": [181], "provides": [182], "highly": [184], "effective": [185], "solution": [186], "next-generation": [188]}, "primary_topic": {"id": "https://openalex.org/T10400", "display_name": "Network Security and Intrusion Detection", "score": 0.9940999746322632, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10400", "display_name": "Network Security and Intrusion Detection", "score": 0.9940999746322632, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11512", "display_name": "Anomaly Detection Techniques and Applications", "score": 0.0013000000035390258, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11598", "display_name": "Internet Traffic Analysis and Secure E-voting", "score": 0.0010000000474974513, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.12732/ijam.v38i11s.1277", "is_oa": true, "landing_page_url": "https://doi.org/10.12732/ijam.v38i11s.1277", "pdf_url": "https://ijamjournal.org/ijam/publication/index.php/ijam/article/download/1277/1166", "source": {"id": "https://openalex.org/S4210189830", "display_name": "International Journal of Apllied Mathematics", "issn_l": "1311-1728", "issn": ["1311-1728", "1314-8060"], "is_oa": true, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "norway-1"], "host_organization": null, "host_organization_name": null, "host_organization_lineage": [], "host_organization_lineage_names": [], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "International Journal of Applied Mathematics", "raw_type": "journal-article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W7117311619", "doi": "https://doi.org/10.48550/arxiv.2512.20920", "title": "RevFFN: Memory-Efficient Full-Parameter Fine-Tuning of Mixture-of-Experts LLMs with Reversible Blocks", "publication_year": 2025, "publication_date": "2025-12-24", "type": "preprint", "language": null, "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5121195460", "display_name": "Ningyuan Liu", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Liu, Ningyuan", "raw_affiliation_strings": ["sen University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "sen University", "institution_ids": []}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5121326548", "display_name": "Jing Yang", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Yang, Jing", "raw_affiliation_strings": ["sen University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "sen University", "institution_ids": []}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5121161992", "display_name": "Kaitong Cai", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Cai, Kaitong", "raw_affiliation_strings": ["sen University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "sen University", "institution_ids": []}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5121316558", "display_name": "Keze Wang", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Wang, Keze", "raw_affiliation_strings": ["sen University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "sen University", "institution_ids": []}]}], "abstract_inverted_index": {"Full": [0], "parameter": [1, 143], "fine": [2, 38, 89, 144, 153], "tuning": [3, 39, 90, 154], "is": [4], "a": [5, 86, 147, 156], "key": [6], "technique": [7], "for": [8, 32, 92, 141], "adapting": [9], "large": [10, 42], "language": [11], "models": [12], "(LLMs)": [13], "to": [14, 24, 27, 119], "downstream": [15], "tasks,": [16], "but": [17, 73], "it": [18], "incurs": [19], "substantial": [20], "memory": [21, 69, 87, 139], "overhead": [22], "due": [23], "the": [25, 117, 128], "need": [26, 118], "cache": [28], "extensive": [29], "intermediate": [30, 122], "activations": [31, 111, 123], "backpropagation.": [33], "This": [34], "bottleneck": [35], "makes": [36], "full": [37, 142, 152], "of": [40, 94, 108, 131], "contemporary": [41], "scale": [43], "LLMs": [44], "challenging": [45], "in": [46, 124], "practice.": [47], "Existing": [48], "distributed": [49], "training": [50, 81], "frameworks": [51], "such": [52], "as": [53], "DeepSpeed": [54], "alleviate": [55], "this": [56, 134], "issue": [57], "using": [58], "techniques": [59], "like": [60], "ZeRO": [61], "and": [62, 79], "FSDP,": [63], "which": [64], "rely": [65], "on": [66, 155], "multi": [67], "GPU": [68], "or": [70, 160], "CPU": [71], "offloading,": [72], "often": [74], "require": [75], "additional": [76], "hardware": [77], "resources": [78], "reduce": [80], "speed.": [82], "We": [83], "introduce": [84], "RevFFN,": [85], "efficient": [88, 151], "paradigm": [91], "mixture": [93], "experts": [95], "(MoE)": [96], "LLMs.": [97], "RevFFN": [98, 149], "employs": [99], "carefully": [100], "designed": [101], "reversible": [102], "Transformer": [103], "blocks": [104], "that": [105], "allow": [106], "reconstruction": [107], "layer": [109], "input": [110], "from": [112], "outputs": [113], "during": [114], "backpropagation,": [115], "eliminating": [116], "store": [120], "most": [121], "memory.": [125], "While": [126], "preserving": [127], "expressive": [129], "capacity": [130], "MoE": [132], "architectures,": [133], "approach": [135], "significantly": [136], "reduces": [137], "peak": [138], "consumption": [140], "tuning.": [145], "As": [146], "result,": [148], "enables": [150], "single": [157], "consumer": [158], "grade": [159, 162], "server": [161], "GPU.": [163]}, "primary_topic": {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.5497999787330627, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.5497999787330627, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11307", "display_name": "Domain Adaptation and Few-Shot Learning", "score": 0.16130000352859497, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.15600000321865082, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2512.20920", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2512.20920", "pdf_url": "https://arxiv.org/pdf/2512.20920", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4416043695", "doi": "https://doi.org/10.48550/arxiv.2502.19119", "title": "Chemical knowledge-informed framework for privacy-aware retrosynthesis learning", "publication_year": 2025, "publication_date": "2025-02-26", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5063404844", "display_name": "Guikun Chen", "orcid": "https://orcid.org/0000-0002-9227-007X", "observed_orcids": ["https://orcid.org/0000-0002-9227-007X"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Chen, Guikun", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5109259463", "display_name": "Xu Zhang", "orcid": "https://orcid.org/0009-0004-8229-810X", "observed_orcids": ["https://orcid.org/0009-0004-8229-810X"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Zhang, Xu", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5103895382", "display_name": "Xiaolin Hu", "orcid": "https://orcid.org/0009-0002-5493-5779", "observed_orcids": ["https://orcid.org/0009-0002-5493-5779"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Hu, Xiaolin", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5028071674", "display_name": "Yong Liu", "orcid": "https://orcid.org/0000-0002-6739-621X", "observed_orcids": ["https://orcid.org/0000-0002-6739-621X"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Liu, Yong", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5005421447", "display_name": "Yi Ping Yang", "orcid": "https://orcid.org/0000-0002-0512-880X", "observed_orcids": ["https://orcid.org/0000-0002-0512-880X"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Yang, Yi", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5101433884", "display_name": "Wenguan Wang", "orcid": "https://orcid.org/0000-0002-0802-9567", "observed_orcids": ["https://orcid.org/0000-0002-0802-9567"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Wang, Wenguan", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Chemical": [0], "reaction": [1, 57, 136, 142, 190], "data": [2, 58, 80, 87], "is": [3], "a": [4, 115, 187, 198], "pivotal": [5], "asset,": [6], "driving": [7], "advances": [8], "in": [9, 40, 176], "competitive": [10, 33], "fields": [11], "such": [12], "as": [13, 26, 76], "pharmaceuticals,": [14], "materials": [15], "science,": [16], "and": [17, 32, 85, 102], "industrial": [18], "chemistry.": [19], "Its": [20], "proprietary": [21, 93, 135], "nature": [22], "renders": [23], "it": [24, 27, 77], "sensitive,": [25], "often": [28], "includes": [29], "confidential": [30], "insights": [31], "advantages": [34], "organizations": [35, 129], "strive": [36], "to": [37, 42, 66, 95, 166], "protect.": [38], "However,": [39], "contrast": [41], "this": [43], "need": [44], "for": [45, 52, 118, 183], "confidentiality,": [46], "the": [47, 105, 110, 132, 158, 169, 179], "current": [48], "standard": [49], "training": [50, 125], "paradigm": [51, 71], "machine": [53], "learning-based": [54], "retrosynthesis": [55, 120, 146], "gathers": [56], "from": [59], "multiple": [60, 127], "sources": [61], "into": [62], "one": [63], "single": [64], "edge": [65], "train": [67], "prediction": [68], "models.": [69, 121], "This": [70], "poses": [72], "considerable": [73], "privacy": [74], "risks": [75], "necessitates": [78], "broad": [79], "availability": [81], "across": [82, 126], "organizational": [83], "boundaries": [84], "frequent": [86], "transmission": [88], "between": [89], "entities,": [90], "potentially": [91], "exposing": [92], "information": [94], "unauthorized": [96], "access": [97], "or": [98], "interception": [99], "during": [100], "storage": [101], "transfer.": [103], "In": [104, 156], "present": [106], "study,": [107], "we": [108], "introduce": [109], "chemical": [111, 128, 150, 159], "knowledge-informed": [112, 151], "framework": [113], "(CKIF),": [114], "privacy-preserving": [116], "approach": [117], "learning": [119], "CKIF": [122, 144, 192], "enables": [123], "distributed": [124], "without": [130], "compromising": [131], "confidentiality": [133], "of": [134, 139, 153, 161, 172, 189], "data.": [137], "Instead": [138], "gathering": [140], "raw": [141], "data,": [143], "learns": [145], "models": [147], "through": [148], "iterative,": [149], "aggregation": [152], "model": [154, 184], "parameters.": [155], "particular,": [157], "properties": [160], "predicted": [162], "reactants": [163], "are": [164], "leveraged": [165], "quantitatively": [167], "assess": [168], "observable": [170], "behaviors": [171], "individual": [173], "models,": [174], "which": [175], "turn": [177], "determines": [178], "adaptive": [180], "weights": [181], "used": [182], "aggregation.": [185], "On": [186], "variety": [188], "datasets,": [191], "outperforms": [193], "several": [194], "strong": [195], "baselines": [196], "by": [197], "clear": [199], "margin.": [200]}, "primary_topic": {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.5981000065803528, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.5981000065803528, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10211", "display_name": "Computational Drug Discovery Methods", "score": 0.2793000042438507, "subfield": {"id": "https://openalex.org/subfields/1703", "display_name": "Computational Theory and Mathematics"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11948", "display_name": "Machine Learning in Materials Science", "score": 0.041999999433755875, "subfield": {"id": "https://openalex.org/subfields/2505", "display_name": "Materials Chemistry"}, "field": {"id": "https://openalex.org/fields/25", "display_name": "Materials Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2502.19119", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2502.19119", "pdf_url": "https://arxiv.org/pdf/2502.19119", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4406798637", "doi": "https://doi.org/10.32628/cseit25111293", "title": "Scaling BERT for Healthcare: An End-to-End Framework for Medical Document Automation", "publication_year": 2025, "publication_date": "2025-01-20", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5116021926", "display_name": "Balamurugan Sivakolunthu Vel", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I118118575", "display_name": "New Jersey Institute of Technology", "ror": "https://ror.org/05e74xb87", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I118118575"]}], "countries": ["US"], "is_corresponding": true, "raw_author_name": "Balamurugan Sivakolunthu Vel", "raw_affiliation_strings": ["New Jersey Institute of Technology"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "New Jersey Institute of Technology", "institution_ids": ["https://openalex.org/I118118575"]}]}], "abstract_inverted_index": {"This": [0, 179], "article": [1], "proposes": [2], "a": [3, 50, 85, 100, 188], "comprehensive": [4], "framework": [5, 83, 131, 181], "for": [6, 13, 43, 191], "implementing": [7, 192], "BERT-based": [8], "language": [9, 194], "models": [10, 26, 195], "at": [11], "scale": [12], "automated": [14], "medical": [15, 35, 63, 93, 102, 150, 205], "document": [16], "processing": [17, 144], "in": [18, 34, 69, 77, 118, 140, 147, 164, 196], "healthcare": [19, 44, 165, 197], "environments.": [20], "Recent": [21], "studies": [22], "have": [23], "demonstrated": [24], "BERT": [25], "achieving": [27], "accuracy": [28, 142], "rates": [29], "of": [30, 62, 172, 204], "up": [31], "to": [32, 57, 90, 108, 183], "89.7%": [33], "entity": [36, 119], "recognition": [37], "tasks": [38], "[2],": [39], "suggesting": [40], "significant": [41], "potential": [42], "applications.": [45], "The": [46, 156], "proposed": [47, 157], "architecture": [48, 159], "introduces": [49], "novel": [51], "hierarchical": [52], "attention": [53], "mechanism": [54], "specifically": [55], "engineered": [56], "capture": [58], "the": [59, 130, 184, 201], "nested": [60], "complexity": [61], "documentation": [64], "while": [65, 113, 199], "maintaining": [66], "computational": [67], "efficiency": [68], "production": [70, 209], "systems.": [71], "Drawing": [72], "inspiration": [73], "from": [74], "successful": [75], "implementations": [76], "pharmacy": [78], "settings": [79, 198], "[15,": [80], "16],": [81], "this": [82], "features": [84], "distributed": [86, 123], "training": [87], "pipeline": [88], "designed": [89], "process": [91, 169], "annotated": [92], "documents": [94, 176], "across": [95], "multiple": [96], "specialties,": [97], "coupled": [98], "with": [99], "dynamic": [101], "vocabulary": [103], "injection": [104], "system": [105, 158], "that": [106], "aims": [107], "preserve": [109], "BERT's": [110], "contextual": [111], "understanding": [112], "potentially": [114], "reducing": [115], "false": [116], "positives": [117], "recognition.": [120], "By": [121], "leveraging": [122], "computing": [124], "infrastructure": [125], "and": [126, 143, 152, 208], "optimized": [127], "model": [128], "architecture,": [129], "could": [132], "theoretically": [133], "achieve": [134], "substantial": [135], "improvements": [136], "over": [137], "traditional": [138], "approaches": [139], "both": [141], "efficiency,": [145], "particularly": [146], "handling": [148], "complex": [149], "terminology": [151], "context-sensitive": [153], "information": [154], "extraction.": [155], "anticipates": [160], "addressing": [161, 200], "current": [162], "challenges": [163, 203], "documentation,": [166], "where": [167], "systems": [168], "an": [170], "average": [171], "6.2": [173], "million": [174], "clinical": [175], "annually": [177], "[1].": [178], "theoretical": [180], "contributes": [182], "field": [185], "by": [186], "proposing": [187], "scalable": [189], "methodology": [190], "advanced": [193], "unique": [202], "domain": [206], "specificity": [207], "deployment": [210], "requirements.": [211]}, "primary_topic": {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.8062000274658203, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.8062000274658203, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T13702", "display_name": "Machine Learning in Healthcare", "score": 0.11209999769926071, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11710", "display_name": "Biomedical Text Mining and Ontologies", "score": 0.05889999866485596, "subfield": {"id": "https://openalex.org/subfields/1312", "display_name": "Molecular Biology"}, "field": {"id": "https://openalex.org/fields/13", "display_name": "Biochemistry, Genetics and Molecular Biology"}, "domain": {"id": "https://openalex.org/domains/1", "display_name": "Life Sciences"}}], "primary_location": {"id": "doi:10.32628/cseit25111293", "is_oa": true, "landing_page_url": "https://doi.org/10.32628/cseit25111293", "pdf_url": "https://ijsrcseit.com/index.php/home/article/download/CSEIT25111293/CSEIT25111293", "source": {"id": "https://openalex.org/S4210200847", "display_name": "International Journal of Scientific Research in Computer Science Engineering and Information Technology", "issn_l": "2456-3307", "issn": ["2456-3307"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": null, "host_organization_name": null, "host_organization_lineage": [], "host_organization_lineage_names": [], "type": "journal"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "International Journal of Scientific Research in Computer Science, Engineering and Information Technology", "raw_type": "journal-article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W7118007540", "doi": "https://doi.org/10.48550/arxiv.2512.24750", "title": "Analyzing Communication Predictability in LLM Training", "publication_year": 2025, "publication_date": "2025-12-31", "type": "preprint", "language": null, "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5019529238", "display_name": "Wenxue Li", "orcid": "https://orcid.org/0000-0002-8228-2552", "observed_orcids": ["https://orcid.org/0000-0002-8228-2552"]}, "institutions": [{"id": "https://openalex.org/I200769079", "display_name": "Hong Kong University of Science and Technology", "ror": "https://ror.org/00q4vv597", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I200769079"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Li, Wenxue", "raw_affiliation_strings": ["Hong Kong University of Science and Technology"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Hong Kong University of Science and Technology", "institution_ids": ["https://openalex.org/I200769079"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5006698160", "display_name": "Xiangzhou Liu", "orcid": "https://orcid.org/0009-0000-2156-9683", "observed_orcids": ["https://orcid.org/0009-0000-2156-9683"]}, "institutions": [{"id": "https://openalex.org/I200769079", "display_name": "Hong Kong University of Science and Technology", "ror": "https://ror.org/00q4vv597", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I200769079"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Liu, Xiangzhou", "raw_affiliation_strings": ["Hong Kong University of Science and Technology"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Hong Kong University of Science and Technology", "institution_ids": ["https://openalex.org/I200769079"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5121790558", "display_name": "Yuxuan Li", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I200769079", "display_name": "Hong Kong University of Science and Technology", "ror": "https://ror.org/00q4vv597", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I200769079"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Li, Yuxuan", "raw_affiliation_strings": ["Hong Kong University of Science and Technology"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Hong Kong University of Science and Technology", "institution_ids": ["https://openalex.org/I200769079"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5087553273", "display_name": "Yilun Jin", "orcid": "https://orcid.org/0000-0002-9502-7622", "observed_orcids": ["https://orcid.org/0000-0002-9502-7622"]}, "institutions": [{"id": "https://openalex.org/I200769079", "display_name": "Hong Kong University of Science and Technology", "ror": "https://ror.org/00q4vv597", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I200769079"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Jin, Yilun", "raw_affiliation_strings": ["Hong Kong University of Science and Technology"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Hong Kong University of Science and Technology", "institution_ids": ["https://openalex.org/I200769079"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5085112687", "display_name": "Zhenghang Ren", "orcid": "https://orcid.org/0000-0002-8779-4768", "observed_orcids": ["https://orcid.org/0000-0002-8779-4768"]}, "institutions": [{"id": "https://openalex.org/I200769079", "display_name": "Hong Kong University of Science and Technology", "ror": "https://ror.org/00q4vv597", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I200769079"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Ren, Zhenghang", "raw_affiliation_strings": ["Hong Kong University of Science and Technology"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Hong Kong University of Science and Technology", "institution_ids": ["https://openalex.org/I200769079"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5052148248", "display_name": "Xudong Liao", "orcid": "https://orcid.org/0000-0002-8380-1879", "observed_orcids": ["https://orcid.org/0000-0002-8380-1879"]}, "institutions": [{"id": "https://openalex.org/I200769079", "display_name": "Hong Kong University of Science and Technology", "ror": "https://ror.org/00q4vv597", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I200769079"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Liao, Xudong", "raw_affiliation_strings": ["Hong Kong University of Science and Technology"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Hong Kong University of Science and Technology", "institution_ids": ["https://openalex.org/I200769079"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5057313730", "display_name": "Han Tian", "orcid": "https://orcid.org/0000-0002-3238-8500", "observed_orcids": ["https://orcid.org/0000-0002-3238-8500"]}, "institutions": [{"id": "https://openalex.org/I200769079", "display_name": "Hong Kong University of Science and Technology", "ror": "https://ror.org/00q4vv597", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I200769079"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Tian, Han", "raw_affiliation_strings": ["Hong Kong University of Science and Technology"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Hong Kong University of Science and Technology", "institution_ids": ["https://openalex.org/I200769079"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5121814732", "display_name": "Bo Ren", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I200769079", "display_name": "Hong Kong University of Science and Technology", "ror": "https://ror.org/00q4vv597", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I200769079"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Ren, Bo", "raw_affiliation_strings": ["Hong Kong University of Science and Technology"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Hong Kong University of Science and Technology", "institution_ids": ["https://openalex.org/I200769079"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5024093604", "display_name": "Zhizhen Zhong", "orcid": "https://orcid.org/0000-0003-3131-374X", "observed_orcids": ["https://orcid.org/0000-0003-3131-374X"]}, "institutions": [{"id": "https://openalex.org/I63966007", "display_name": "Massachusetts Institute of Technology", "ror": "https://ror.org/042nb2s44", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I63966007"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Zhong, Zhizhen", "raw_affiliation_strings": ["Massachusetts Institute of Technology"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Massachusetts Institute of Technology", "institution_ids": ["https://openalex.org/I63966007"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5065693468", "display_name": "Guyue Liu", "orcid": "https://orcid.org/0009-0001-5822-3835", "observed_orcids": ["https://orcid.org/0009-0001-5822-3835", "https://orcid.org/0009-0009-0595-2127", "https://orcid.org/0000-0002-5103-3174"]}, "institutions": [{"id": "https://openalex.org/I20231570", "display_name": "Peking University", "ror": "https://ror.org/02v51f717", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I20231570"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Liu, Guyue", "raw_affiliation_strings": ["Peking University"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Peking University", "institution_ids": ["https://openalex.org/I20231570"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5121828483", "display_name": "Ying Zhang", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Zhang, Ying", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5121820035", "display_name": "Kai Chen", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I200769079", "display_name": "Hong Kong University of Science and Technology", "ror": "https://ror.org/00q4vv597", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I200769079"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Chen, Kai", "raw_affiliation_strings": ["Hong Kong University of Science and Technology"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Hong Kong University of Science and Technology", "institution_ids": ["https://openalex.org/I200769079"]}]}], "abstract_inverted_index": {"Effective": [0], "communication": [1, 44, 67, 93, 103], "is": [2, 109], "essential": [3], "in": [4, 46, 50, 75, 105, 146], "distributed": [5, 47], "training,": [6, 48, 107], "with": [7, 111], "predictability": [8, 23, 45], "being": [9], "one": [10], "of": [11, 34], "its": [12], "most": [13], "significant": [14], "characteristics.": [15], "However,": [16], "existing": [17], "studies": [18], "primarily": [19], "focus": [20], "on": [21, 62], "exploiting": [22], "through": [24], "online": [25], "profiling": [26], "for": [27], "runtime": [28], "optimization,": [29], "without": [30], "a": [31, 122, 143], "systematic": [32], "understanding": [33], "it.": [35], "In": [36], "this": [37, 118], "work,": [38], "we": [39, 70, 96, 120], "aim": [40], "to": [41, 101, 127, 132, 142, 149], "systematically": [42], "formulate": [43], "particularly": [49], "Large": [51], "Language": [52], "Models": [53], "(LLMs)": [54], "that": [55], "utilize": [56], "hybrid": [57], "parallelism.": [58], "Our": [59], "analysis": [60], "focuses": [61], "both": [63], "traffic": [64, 73], "patterns": [65, 74], "and": [66, 78, 86], "overhead.": [68], "Specifically,": [69], "investigate": [71], "predictable": [72], "typical": [76], "LLMs": [77], "evaluate": [79], "how": [80], "various": [81], "factors": [82], "influence": [83], "GPU": [84], "utilization": [85], "effective": [87], "bandwidth": [88], "(two": [89], "critical": [90], "variables": [91], "affecting": [92], "overhead).": [94], "Furthermore,": [95], "develop": [97], "an": [98], "analytical": [99], "formulation": [100], "estimate": [102], "overhead": [104], "LLM": [106], "which": [108], "validated": [110], "high": [112], "accuracy": [113], "against": [114], "empirical": [115], "data.": [116], "Leveraging": [117], "formulation,": [119], "propose": [121], "configuration": [123, 155], "tuning": [124], "tool,": [125], "ConfigTuner,": [126], "optimize": [128], "training": [129, 135], "performance.": [130], "Compared": [131, 148], "Megatron-LM,": [133], "the": [134, 153, 160], "configurations": [136], "optimized": [137], "by": [138], "ConfigTuner": [139, 151], "demonstrate": [140], "up": [141], "1.36$\\times$": [144], "increase": [145], "throughput.": [147], "Alpa,": [150], "generates": [152], "same": [154], "suggestion": [156], "while": [157], "significantly": [158], "reducing": [159], "search": [161], "complexity.": [162]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9603999853134155, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9603999853134155, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10715", "display_name": "Distributed and Parallel Computing Systems", "score": 0.0052999998442828655, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.00419999985024333, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2512.24750", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2512.24750", "pdf_url": "https://arxiv.org/pdf/2512.24750", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4415199741", "doi": "https://doi.org/10.48550/arxiv.2505.10183", "title": "KAITIAN: A Unified Communication Framework for Enabling Efficient Collaboration Across Heterogeneous Accelerators in Embodied AI Systems", "publication_year": 2025, "publication_date": "2025-05-15", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Lin, Jieke", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I4210165038", "display_name": "University of Chinese Academy of Sciences", "ror": "https://ror.org/05qbk4x57", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I19820366", "https://openalex.org/I4210165038"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Lin, Jieke", "raw_affiliation_strings": ["Hangzhou Institute for Advanced Study , UCAS"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Hangzhou Institute for Advanced Study , UCAS", "institution_ids": ["https://openalex.org/I4210165038"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Wang, Wanyu", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I4210165038", "display_name": "University of Chinese Academy of Sciences", "ror": "https://ror.org/05qbk4x57", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I19820366", "https://openalex.org/I4210165038"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Wang, Wanyu", "raw_affiliation_strings": ["Hangzhou Institute for Advanced Study , UCAS"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Hangzhou Institute for Advanced Study , UCAS", "institution_ids": ["https://openalex.org/I4210165038"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5057301818", "display_name": "Longxiang Yin", "orcid": "https://orcid.org/0000-0003-2924-3949", "observed_orcids": ["https://orcid.org/0000-0003-2924-3949"]}, "institutions": [{"id": "https://openalex.org/I19820366", "display_name": "Chinese Academy of Sciences", "ror": "https://ror.org/034t30j35", "country_code": "CN", "type": "government", "lineage": ["https://openalex.org/I19820366"]}, {"id": "https://openalex.org/I4210090176", "display_name": "Institute of Computing Technology", "ror": "https://ror.org/0090r4d87", "country_code": "CN", "type": "facility", "lineage": ["https://openalex.org/I19820366", "https://openalex.org/I4210090176"]}, {"id": "https://openalex.org/I4210165038", "display_name": "University of Chinese Academy of Sciences", "ror": "https://ror.org/05qbk4x57", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I19820366", "https://openalex.org/I4210165038"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yin, Longxiang", "raw_affiliation_strings": ["Hangzhou Institute for Advanced Study , UCAS", "Institute of Computing Technology , Chinese Academy of Sciences , Beijing , China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Hangzhou Institute for Advanced Study , UCAS", "institution_ids": ["https://openalex.org/I4210165038"]}, {"raw_affiliation_string": "Institute of Computing Technology , Chinese Academy of Sciences , Beijing , China", "institution_ids": ["https://openalex.org/I19820366", "https://openalex.org/I4210090176"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5101512833", "display_name": "Yinhe Han", "orcid": "https://orcid.org/0000-0001-5113-8067", "observed_orcids": ["https://orcid.org/0000-0001-5113-8067", "https://orcid.org/0000-0003-0904-6681"]}, "institutions": [{"id": "https://openalex.org/I4210165038", "display_name": "University of Chinese Academy of Sciences", "ror": "https://ror.org/05qbk4x57", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I19820366", "https://openalex.org/I4210165038"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Han, Yinhe", "raw_affiliation_strings": ["Hangzhou Institute for Advanced Study , UCAS"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Hangzhou Institute for Advanced Study , UCAS", "institution_ids": ["https://openalex.org/I4210165038"]}]}], "abstract_inverted_index": {"Embodied": [0], "Artificial": [1], "Intelligence": [2], "(AI)": [3], "systems,": [4, 168], "such": [5], "as": [6, 121], "autonomous": [7], "robots": [8], "and": [9, 28, 50, 56, 126, 135, 145, 175, 186], "intelligent": [10], "vehicles,": [11], "are": [12], "increasingly": [13], "reliant": [14], "on": [15, 115, 129], "diverse": [16], "heterogeneous": [17, 112, 188], "accelerators": [18], "(e.g.,": [19], "GPGPUs,": [20], "NPUs,": [21], "FPGAs)": [22], "to": [23, 52, 73, 124, 162, 165], "meet": [24], "stringent": [25], "real-time": [26, 117], "processing": [27], "energy-efficiency": [29], "demands.": [30], "However,": [31], "the": [32, 181], "proliferation": [33], "of": [34], "vendor-specific": [35], "proprietary": [36], "communication": [37, 70, 87, 94, 172], "libraries": [38, 88], "creates": [39], "significant": [40, 140], "interoperability": [41], "barriers,": [42], "hindering": [43], "seamless": [44], "collaboration": [45], "between": [46], "different": [47], "accelerator": [48], "types": [49], "leading": [51], "suboptimal": [53], "resource": [54, 143], "utilization": [55, 144], "performance": [57, 118], "bottlenecks": [58], "in": [59, 142, 190], "distributed": [60, 69, 148], "AI": [61, 193], "workloads.": [62], "This": [63], "paper": [64], "introduces": [65], "KAITIAN,": [66], "a": [67, 79, 102, 130], "novel": [68], "framework": [71], "designed": [72], "bridge": [74], "this": [75], "gap.": [76], "KAITIAN": [77, 138, 155, 179], "provides": [78], "unified": [80], "abstraction": [81], "layer": [82], "that": [83, 106, 154], "intelligently": [84], "integrates": [85], "vendor-optimized": [86], "for": [89, 96, 147, 183], "intra-group": [90], "efficiency": [91], "with": [92], "general-purpose": [93], "protocols": [95], "inter-group": [97], "interoperability.": [98], "Crucially,": [99], "it": [100], "incorporates": [101], "load-adaptive": [103], "scheduling": [104], "mechanism": [105], "dynamically": [107], "balances": [108], "computational": [109], "tasks": [110], "across": [111], "devices": [113], "based": [114], "their": [116], "characteristics.": [119], "Implemented": [120], "an": [122], "extension": [123], "PyTorch": [125], "rigorously": [127], "evaluated": [128], "testbed": [131], "featuring": [132], "NVIDIA": [133], "GPUs": [134], "Cambricon": [136], "MLUs,": [137], "demonstrates": [139], "improvements": [141], "scalability": [146], "training": [149, 158], "tasks.": [150], "Experimental": [151], "results": [152], "show": [153], "can": [156], "accelerate": [157], "time": [159], "by": [160], "up": [161], "42%": [163], "compared": [164], "baseline": [166], "homogeneous": [167], "while": [169], "incurring": [170], "minimal": [171], "overhead": [173], "(2.8--4.3%)": [174], "maintaining": [176], "model": [177], "accuracy.": [178], "paves": [180], "way": [182], "more": [184], "flexible": [185], "powerful": [187], "computing": [189], "complex": [191], "embodied": [192], "applications.": [194]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9794999957084656, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9794999957084656, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10715", "display_name": "Distributed and Parallel Computing Systems", "score": 0.006300000008195639, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.0017000000225380063, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2505.10183", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2505.10183", "pdf_url": "https://arxiv.org/pdf/2505.10183", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 1, "is_retracted": false}, {"id": "https://openalex.org/W7117256254", "doi": "https://doi.org/10.48550/arxiv.2512.20064", "title": "FastMPS: Revisit Data Parallel in Large-scale Matrix Product State Sampling", "publication_year": 2025, "publication_date": "2025-12-23", "type": "preprint", "language": null, "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5033807671", "display_name": "Yaojian Chen", "orcid": "https://orcid.org/0000-0002-6363-134X", "observed_orcids": ["https://orcid.org/0000-0002-6363-134X"]}, "institutions": [{"id": "https://openalex.org/I99065089", "display_name": "Tsinghua University", "ror": "https://ror.org/03cve4549", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I99065089"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Chen, Yaojian", "raw_affiliation_strings": ["Department of Computer Science and Technology , Tsinghua University Beijing , China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Computer Science and Technology , Tsinghua University Beijing , China", "institution_ids": ["https://openalex.org/I99065089"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5074767359", "display_name": "Si-Qiu Gong", "orcid": "https://orcid.org/0000-0002-5447-2228", "observed_orcids": ["https://orcid.org/0000-0002-5447-2228"]}, "institutions": [{"id": "https://openalex.org/I126520041", "display_name": "University of Science and Technology of China", "ror": "https://ror.org/04c4dkn09", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I126520041", "https://openalex.org/I19820366"]}, {"id": "https://openalex.org/I4210103894", "display_name": "Hefei National Center for Physical Sciences at Nanoscale", "ror": "https://ror.org/01jeedh73", "country_code": "CN", "type": "facility", "lineage": ["https://openalex.org/I126520041", "https://openalex.org/I19820366", "https://openalex.org/I4210103894"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Gong, Si-Qiu", "raw_affiliation_strings": ["Hefei National Research Center for Physical Sciences at the Microscale and", "School of Physical Sciences , University of Science and Technology of China Hefei , China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Hefei National Research Center for Physical Sciences at the Microscale and", "institution_ids": ["https://openalex.org/I4210103894"]}, {"raw_affiliation_string": "School of Physical Sciences , University of Science and Technology of China Hefei , China", "institution_ids": ["https://openalex.org/I126520041"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5121261536", "display_name": "Lin Gan", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I99065089", "display_name": "Tsinghua University", "ror": "https://ror.org/03cve4549", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I99065089"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Gan, Lin", "raw_affiliation_strings": ["Department of Computer Science and Technology , Tsinghua University Beijing , China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Computer Science and Technology , Tsinghua University Beijing , China", "institution_ids": ["https://openalex.org/I99065089"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5121274756", "display_name": "Yanfei Liu", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I4210158984", "display_name": "National Supercomputing Center in Wuxi", "ror": "https://ror.org/04ypjrs34", "country_code": "CN", "type": "other", "lineage": ["https://openalex.org/I4210158984"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Liu, Yanfei", "raw_affiliation_strings": ["National Supercomputing Center in Wuxi Wuxi , China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "National Supercomputing Center in Wuxi Wuxi , China", "institution_ids": ["https://openalex.org/I4210158984"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5121274394", "display_name": "An Yang", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I99065089", "display_name": "Tsinghua University", "ror": "https://ror.org/03cve4549", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I99065089"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yang, An", "raw_affiliation_strings": ["Department of Computer Science and Technology , Tsinghua University Beijing , China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Computer Science and Technology , Tsinghua University Beijing , China", "institution_ids": ["https://openalex.org/I99065089"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5121279345", "display_name": "Yinuo Wang", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I99065089", "display_name": "Tsinghua University", "ror": "https://ror.org/03cve4549", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I99065089"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Wang, Yinuo", "raw_affiliation_strings": ["Department of Computer Science and Technology , Tsinghua University Beijing , China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Computer Science and Technology , Tsinghua University Beijing , China", "institution_ids": ["https://openalex.org/I99065089"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5121283097", "display_name": "Chao-yang Lu", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I126520041", "display_name": "University of Science and Technology of China", "ror": "https://ror.org/04c4dkn09", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I126520041", "https://openalex.org/I19820366"]}, {"id": "https://openalex.org/I4210103894", "display_name": "Hefei National Center for Physical Sciences at Nanoscale", "ror": "https://ror.org/01jeedh73", "country_code": "CN", "type": "facility", "lineage": ["https://openalex.org/I126520041", "https://openalex.org/I19820366", "https://openalex.org/I4210103894"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Lu, Chao-yang", "raw_affiliation_strings": ["Hefei National Research Center for Physical Sciences at the Microscale and", "School of Physical Sciences , University of Science and Technology of China Hefei , China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Hefei National Research Center for Physical Sciences at the Microscale and", "institution_ids": ["https://openalex.org/I4210103894"]}, {"raw_affiliation_string": "School of Physical Sciences , University of Science and Technology of China Hefei , China", "institution_ids": ["https://openalex.org/I126520041"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5121273406", "display_name": "Guangwen Yang", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I99065089", "display_name": "Tsinghua University", "ror": "https://ror.org/03cve4549", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I99065089"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yang, Guangwen", "raw_affiliation_strings": ["Department of Computer Science and Technology , Tsinghua University Beijing , China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Computer Science and Technology , Tsinghua University Beijing , China", "institution_ids": ["https://openalex.org/I99065089"]}]}], "abstract_inverted_index": {"Matrix": [0], "Product": [1], "State": [2], "(MPS)": [3], "is": [4, 42, 48], "a": [5, 25, 76, 123], "versatile": [6], "tensor": [7, 92, 168], "network": [8, 169], "representation": [9], "widely": [10], "applied": [11], "in": [12, 29, 55, 111, 166], "quantum": [13, 15], "physics,": [14], "chemistry,": [16], "and": [17, 52, 69, 100, 105, 107, 125, 142, 148], "machine": [18], "learning,": [19], "etc.": [20], "MPS": [21, 41, 64, 82, 113], "sampling": [22], "serves": [23], "as": [24], "critical": [26], "fundamental": [27], "operation": [28], "these": [30], "fields.": [31], "As": [32], "the": [33, 38, 156, 159], "problems": [34], "become": [35], "more": [36], "complex,": [37], "scale": [39], "of": [40, 140, 158], "rapidly": [43], "increasing.": [44], "Traditional": [45], "data": [46, 87, 109], "parallelism": [47, 59, 88, 93], "limited": [49], "by": [50], "memory": [51, 99], "heavy": [53], "I/O": [54, 101], "large-scale": [56, 63, 112], "MPS.": [57], "Model": [58], "that": [60], "can": [61], "handle": [62], "imposes": [65], "rigid": [66], "process": [67], "bindings": [68], "lacks": [70], "scalability.": [71], "This": [72], "work": [73], "proposes": [74], "Fast-MPS,": [75], "multi-level": [77], "parallel": [78, 110], "framework": [79], "for": [80], "scalable": [81], "sampling.": [83, 114], "Our": [84], "design": [85], "combines": [86], "across": [89], "samples": [90], "with": [91, 145], "along": [94], "bond": [95, 149], "dimensions.": [96], "We": [97, 115], "eliminate": [98], "pressure": [102], "through": [103], "compression": [104], "overlapping,": [106], "revive": [108], "evaluate": [116], "our": [117], "approach": [118], "on": [119], "Gaussian": [120], "Boson": [121], "Sampling,": [122], "representative": [124], "demanding": [126], "application.": [127], "Fast-MPS": [128, 161], "achieves": [129], "over": [130], "10x": [131], "speedup": [132], "compared": [133], "to": [134, 138], "existing": [135], "simulators,": [136], "scales": [137], "thousands": [139], "processes,": [141], "enables": [143], "simulations": [144], "8,176": [146], "sites": [147], "dimension": [150], "chi": [151], "=": [152], "10^4,": [153], "significantly": [154], "outperforming": [155], "state": [157], "art.": [160], "has": [162], "demonstrated": [163], "great": [164], "potential": [165], "high-performance": [167], "applications.": [170]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.454800009727478, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.454800009727478, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10682", "display_name": "Quantum Computing Algorithms and Architecture", "score": 0.4004000127315521, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11804", "display_name": "Quantum many-body systems", "score": 0.10610000044107437, "subfield": {"id": "https://openalex.org/subfields/3107", "display_name": "Atomic and Molecular Physics, and Optics"}, "field": {"id": "https://openalex.org/fields/31", "display_name": "Physics and Astronomy"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2512.20064", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2512.20064", "pdf_url": "https://arxiv.org/pdf/2512.20064", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4414977014", "doi": "https://doi.org/10.48550/arxiv.2510.05468", "title": "AMAQ: Adaptive Mixed-bit Activation Quantization for Collaborative Parameter Efficient Fine-tuning", "publication_year": 2025, "publication_date": "2025-10-07", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5082442045", "display_name": "Yurun Song", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Song, Yurun", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5023011245", "display_name": "Zhuoyi Yang", "orcid": "https://orcid.org/0009-0008-7909-4930", "observed_orcids": ["https://orcid.org/0009-0008-7909-4930"]}, "institutions": [{"id": "https://openalex.org/I204250578", "display_name": "University of California, Irvine", "ror": "https://ror.org/04gyf1771", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I204250578"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Yang, Zhuoyi", "raw_affiliation_strings": ["UC Irvine"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "UC Irvine", "institution_ids": ["https://openalex.org/I204250578"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5037855756", "display_name": "Ian G. Harris", "orcid": "https://orcid.org/0000-0003-3244-6528", "observed_orcids": ["https://orcid.org/0000-0003-3244-6528"]}, "institutions": [{"id": "https://openalex.org/I204250578", "display_name": "University of California, Irvine", "ror": "https://ror.org/04gyf1771", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I204250578"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Harris, Ian G.", "raw_affiliation_strings": ["UC Irvine"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "UC Irvine", "institution_ids": ["https://openalex.org/I204250578"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5058967466", "display_name": "Sangeetha Abdu Jyothi", "orcid": "https://orcid.org/0009-0000-0503-4478", "observed_orcids": ["https://orcid.org/0009-0000-0503-4478"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Jyothi, Sangeetha Abdu", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Large": [0], "Language": [1], "Models": [2], "(LLMs)": [3], "are": [4], "scaling": [5], "rapidly,": [6], "creating": [7], "significant": [8], "challenges": [9], "for": [10, 40, 125, 171, 186], "collaborative": [11, 41, 51, 158, 187], "server": [12], "client": [13], "distributed": [14], "training,": [15, 52], "particularly": [16], "in": [17, 50], "terms": [18], "of": [19], "communication": [20, 48, 169, 191], "efficiency": [21, 37], "and": [22, 38, 67, 97, 119, 130, 140, 183], "computational": [23], "overheads.": [24], "To": [25, 46], "address": [26], "these": [27], "challenges,": [28], "we": [29, 53], "implement": [30], "Parameter-efficient": [31], "Split": [32], "Learning,": [33], "which": [34], "effectively": [35, 87, 154], "balances": [36], "performance": [39], "training": [42, 138, 159, 188], "on": [43, 94], "low-resource": [44], "devices.": [45], "reduce": [47], "overhead": [49, 170], "introduce": [54], "Adaptive": [55], "Mixed": [56], "bit": [57, 89, 102, 107, 143], "Activation": [58], "Quantization": [59], "(AMAQ),": [60], "a": [61, 167, 181], "strategy": [62], "that": [63, 151], "progressively": [64], "compresses": [65], "activations": [66], "gradients": [68], "from": [69], "high": [70], "precision": [71, 78], "(6": [72], "to": [73, 76, 80], "8": [74], "bits)": [75], "low": [77], "(3": [79], "4": [81], "bits).": [82], "AMAQ": [83, 109, 152, 180], "achieves": [84], "this": [85], "by": [86], "allocating": [88], "budgets": [90], "across": [91], "channels": [92], "based": [93], "feature": [95], "wise": [96, 99], "layer": [98], "importance": [100], "using": [101], "regularization.": [103], "Under": [104], "the": [105, 147], "same": [106], "budgets,": [108], "outperforms": [110], "fixed-precision": [111], "approaches,": [112], "delivering": [113], "about": [114, 120], "2.5%": [115], "higher": [116], "generation": [117], "accuracy": [118, 124, 164], "1.3%": [121], "better": [122], "classification": [123], "models": [126], "like": [127], "LLaMA3": [128], "8B": [129], "Qwen2.5": [131], "7B.": [132], "In": [133], "addition,": [134], "it": [135], "significantly": [136], "enhances": [137], "stability": [139], "reducing": [141], "ultra-low": [142], "representation": [144], "collapse": [145], "during": [146, 174], "training.": [148, 175], "Experiments": [149], "demonstrate": [150], "integrates": [153], "into": [155], "practical": [156, 182], "multi-machine": [157], "setups,": [160], "offering": [161], "superior": [162], "inference": [163], "with": [165, 189], "only": [166], "modest": [168], "bits": [172], "adaptation": [173], "This": [176], "trade": [177], "off": [178], "makes": [179], "effective": [184], "solution": [185], "minimal": [190], "cost.": [192]}, "primary_topic": {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.8906000256538391, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.8906000256538391, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11307", "display_name": "Domain Adaptation and Few-Shot Learning", "score": 0.0203000009059906, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.019500000402331352, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2510.05468", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2510.05468", "pdf_url": "https://arxiv.org/pdf/2510.05468", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4406316662", "doi": "https://doi.org/10.48550/arxiv.2501.05563", "title": "Prediction-Assisted Online Distributed Deep Learning Workload Scheduling in GPU Clusters", "publication_year": 2025, "publication_date": "2025-01-09", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5048036564", "display_name": "Ziyue Luo", "orcid": "https://orcid.org/0000-0002-1253-8137", "observed_orcids": ["https://orcid.org/0000-0002-1253-8137"]}, "institutions": [{"id": "https://openalex.org/I52357470", "display_name": "The Ohio State University", "ror": "https://ror.org/00rs6vg23", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I52357470"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Luo, Ziyue", "raw_affiliation_strings": ["Dept. of ECE , The Ohio State University , USA,"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Dept. of ECE , The Ohio State University , USA,", "institution_ids": ["https://openalex.org/I52357470"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5103194402", "display_name": "Jia Liu", "orcid": "https://orcid.org/0009-0002-3971-9621", "observed_orcids": ["https://orcid.org/0009-0002-3971-9621"]}, "institutions": [{"id": "https://openalex.org/I52357470", "display_name": "The Ohio State University", "ror": "https://ror.org/00rs6vg23", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I52357470"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Liu, Jia", "raw_affiliation_strings": ["Dept. of ECE , The Ohio State University , USA,"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Dept. of ECE , The Ohio State University , USA,", "institution_ids": ["https://openalex.org/I52357470"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5101640754", "display_name": "Myungjin Lee", "orcid": "https://orcid.org/0000-0002-0246-9676", "observed_orcids": ["https://orcid.org/0000-0002-0246-9676"]}, "institutions": [{"id": "https://openalex.org/I135428043", "display_name": "Cisco Systems (United States)", "ror": "https://ror.org/03yt1ez60", "country_code": "US", "type": "company", "lineage": ["https://openalex.org/I135428043"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Lee, Myungjin", "raw_affiliation_strings": ["Cisco Research , USA ,"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Cisco Research , USA ,", "institution_ids": ["https://openalex.org/I135428043"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5035752536", "display_name": "Ness B. Shroff", "orcid": "https://orcid.org/0000-0002-4606-6879", "observed_orcids": ["https://orcid.org/0000-0002-4606-6879"]}, "institutions": [{"id": "https://openalex.org/I52357470", "display_name": "The Ohio State University", "ror": "https://ror.org/00rs6vg23", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I52357470"]}], "countries": ["US"], "is_corresponding": false, "raw_author_name": "Shroff, Ness B.", "raw_affiliation_strings": ["Dept. of ECE , The Ohio State University , USA,"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Dept. of ECE , The Ohio State University , USA,", "institution_ids": ["https://openalex.org/I52357470"]}]}], "abstract_inverted_index": {"The": [0], "recent": [1], "explosive": [2], "growth": [3], "of": [4], "deep": [5, 20], "learning": [6, 21], "(DL)": [7], "models": [8, 69], "has": [9], "necessitated": [10], "a": [11, 39, 60, 97, 114, 122, 131, 143], "compelling": [12], "need": [13], "for": [14, 18, 133], "efficient": [15], "job": [16, 58, 135], "scheduling": [17, 37, 43, 111, 136, 147], "distributed": [19, 73], "training": [22, 74, 104], "with": [23, 51], "mixed": [24], "parallelisms": [25], "(DDLwMP)": [26], "in": [27], "GPU": [28, 139], "clusters.": [29], "This": [30, 126], "paper": [31], "proposes": [32], "an": [33], "adaptive": [34], "shortest-remaining-processing-time-first": [35], "(A-SRPT)": [36], "algorithm,": [38], "novel": [40], "prediction-assisted": [41], "online": [42], "approach": [44], "designed": [45], "to": [46, 63, 80, 102, 142, 157], "mitigate": [47], "the": [48, 81, 109, 138], "challenges": [49], "associated": [50, 72], "DL": [52], "cluster": [53], "scheduling.": [54], "By": [55], "modeling": [56], "each": [57], "as": [59, 130], "graph": [61], "corresponding": [62], "heterogeneous": [64], "Deep": [65], "Neural": [66], "Network": [67], "(DNN)": [68], "and": [70, 154], "their": [71], "configurations,": [75], "A-SRPT": [76, 95, 107], "strategically": [77], "assigns": [78], "jobs": [79, 93], "available": [82], "GPUs,": [83], "thereby": [84], "minimizing": [85], "inter-server": [86], "communication": [87], "overhead.": [88], "Observing": [89], "that": [90], "most": [91], "DDLwMP": [92], "recur,": [94], "incorporates": [96], "random": [98], "forest": [99], "regression": [100], "model": [101], "predict": [103], "iterations.": [105], "Crucially,": [106], "maps": [108], "complex": [110], "problem": [112], "into": [113], "single-machine": [115], "instance,": [116], "which": [117], "is": [118], "addressed": [119], "optimally": [120], "by": [121], "preemptive": [123], "\"shortest-remaining-processing-time-first\"": [124], "strategy.": [125], "optimized": [127], "solution": [128], "serves": [129], "guide": [132], "actual": [134], "within": [137], "clusters,": [140], "leading": [141], "theoretically": [144], "provable": [145], "competitive": [146], "efficiency.": [148], "We": [149], "conduct": [150], "extensive": [151], "real-world": [152], "testbed": [153], "simulation": [155], "experiments": [156], "verify": [158], "our": [159], "proposed": [160], "algorithms.": [161]}, "primary_topic": {"id": "https://openalex.org/T10715", "display_name": "Distributed and Parallel Computing Systems", "score": 0.5971999764442444, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10715", "display_name": "Distributed and Parallel Computing Systems", "score": 0.5971999764442444, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.2152000069618225, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.12489999830722809, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2501.05563", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2501.05563", "pdf_url": "https://arxiv.org/pdf/2501.05563", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 1, "is_retracted": false}, {"id": "https://openalex.org/W4417452288", "doi": "https://doi.org/10.48550/arxiv.2512.12801", "title": "Fine-Grained Energy Prediction For Parallellized LLM Inference With PIE-P", "publication_year": 2025, "publication_date": "2025-12-14", "type": "preprint", "language": null, "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5009447097", "display_name": "Anurag Dutt", "orcid": "https://orcid.org/0009-0009-9445-3214", "observed_orcids": ["https://orcid.org/0009-0009-9445-3214"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Dutt, Anurag", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": null, "display_name": "Choi, Young Won", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Choi, Young Won", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5112335036", "display_name": "Avirup Sil", "orcid": "https://orcid.org/0000-0002-4753-3221", "observed_orcids": ["https://orcid.org/0000-0002-4753-3221"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Sil, Avirup", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5024346889", "display_name": "Anshul Gandhi", "orcid": "https://orcid.org/0000-0002-2351-6523", "observed_orcids": ["https://orcid.org/0000-0002-2351-6523"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Gandhi, Anshul", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5048867459", "display_name": "Aruna Balasubramanian", "orcid": "https://orcid.org/0000-0003-3720-2215", "observed_orcids": ["https://orcid.org/0000-0003-3720-2215"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Balasubramanian, Aruna", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5101768349", "display_name": "Niranjan Balasubramanian", "orcid": "https://orcid.org/0000-0003-4187-9368", "observed_orcids": ["https://orcid.org/0000-0003-4187-9368"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Balasubramanian, Niranjan", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"With": [0], "the": [1, 23, 107, 115, 129], "widespread": [2], "adoption": [3], "of": [4, 11, 26, 147, 153], "Large": [5], "Language": [6], "Models": [7], "(LLMs),": [8], "energy": [9, 24, 41, 57, 94, 108, 127, 166], "costs": [10], "running": [12], "LLMs": [13, 27], "is": [14, 28, 76, 112], "quickly": [15], "becoming": [16], "a": [17, 92, 134], "critical": [18], "concern.": [19], "However,": [20], "precisely": [21], "measuring": [22], "consumption": [25], "often": [29], "infeasible": [30], "because": [31], "hardware-based": [32], "power": [33], "monitors": [34], "are": [35, 44, 61, 68], "not": [36, 45, 69], "always": [37], "accessible": [38], "and": [39, 66, 89, 103, 123, 150, 164], "software-based": [40], "measurement": [42], "tools": [43], "accurate.": [46], "While": [47], "various": [48], "prediction": [49, 95, 136], "techniques": [50], "have": [51], "been": [52], "developed": [53], "to": [54, 63, 71], "estimate": [55], "LLM": [56, 73], "consumption,": [58], "these": [59, 140], "approaches": [60], "limited": [62], "single-GPU": [64], "environments": [65], "thus": [67], "applicable": [70], "modern": [72], "inference": [74, 111], "which": [75], "typically": [77], "parallelized": [78, 110], "across": [79, 168], "multiple": [80], "GPUs.": [81], "In": [82], "this": [83, 87], "work,": [84], "we": [85], "remedy": [86], "gap": [88], "introduce": [90], "PIE-P,": [91], "fine-grained": [93, 145, 165], "framework": [96, 137], "for": [97], "multi-GPU": [98], "inference,": [99], "including": [100], "tensor,": [101], "pipeline,": [102], "data": [104], "parallelism.": [105], "Predicting": [106], "under": [109], "complicated": [113], "by": [114], "non-determinism": [116], "in": [117, 125], "inter-GPU": [118, 148], "communication,": [119, 149], "additional": [120], "communication": [121], "overheads,": [122], "difficulties": [124], "isolating": [126], "during": [128], "communication/synchronization": [130], "phase.": [131], "We": [132], "develop": [133], "scalable": [135], "that": [138, 160], "addresses": [139], "issues": [141], "via": [142], "precise": [143], "sampling,": [144], "modeling": [146], "careful": [151], "accounting": [152], "parallelization": [154], "overhead.": [155], "Our": [156], "evaluation": [157], "results": [158], "show": [159], "PIE-P": [161], "yields": [162], "accurate": [163], "predictions": [167], "parallelism": [169], "strategies,": [170], "significantly": [171], "outperforming": [172], "baselines.": [173]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.8974999785423279, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.8974999785423279, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T14098", "display_name": "Energy Efficiency in Computing", "score": 0.04809999838471413, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T12238", "display_name": "Green IT and Sustainability", "score": 0.0414000004529953, "subfield": {"id": "https://openalex.org/subfields/2208", "display_name": "Electrical and Electronic Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2512.12801", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2512.12801", "pdf_url": "https://arxiv.org/pdf/2512.12801", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4414384331", "doi": "https://doi.org/10.48550/arxiv.2504.06439", "title": "Graph Neural Network-Based Distributed Optimal Control for Linear Networked Systems: An Online Distributed Training Approach", "publication_year": 2025, "publication_date": "2025-04-08", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5109119591", "display_name": "Zihao Song", "orcid": "https://orcid.org/0000-0001-6872-6164", "observed_orcids": ["https://orcid.org/0000-0001-6872-6164", "https://orcid.org/0009-0004-6859-7090", "https://orcid.org/0009-0009-2593-4460"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Song, Zihao", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5074601746", "display_name": "Shirantha Welikala", "orcid": "https://orcid.org/0000-0003-3470-7082", "observed_orcids": ["https://orcid.org/0000-0003-3470-7082"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Welikala, Shirantha", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5077171773", "display_name": "Panos J. Antsaklis", "orcid": "https://orcid.org/0000-0002-6836-5028", "observed_orcids": ["https://orcid.org/0000-0002-6836-5028"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Antsaklis, Panos J.", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5045445273", "display_name": "Hai Lin", "orcid": "https://orcid.org/0000-0002-5242-2366", "observed_orcids": ["https://orcid.org/0000-0002-5242-2366"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Lin, Hai", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"In": [0, 15], "this": [1, 84], "paper,": [2], "we": [3, 17, 86, 96], "consider": [4], "the": [5, 33, 47, 53, 76, 98, 106, 119, 132, 137, 151, 156], "distributed": [6, 22, 71, 91, 107, 113, 121, 125], "optimal": [7, 23, 39, 54, 92], "control": [8, 93], "problem": [9, 99], "for": [10], "discrete-time": [11], "linear": [12, 138], "networked": [13, 139], "systems.": [14], "particular,": [16], "are": [18, 56, 63, 74], "interested": [19], "in": [20, 37, 68], "learning": [21, 103], "controllers": [24, 40, 55], "using": [25, 176], "graph": [26], "recurrent": [27], "neural": [28], "networks": [29], "(GRNNs).": [30], "Most": [31], "of": [32, 50, 79, 136, 155, 167], "existing": [34], "approaches": [35], "result": [36], "centralized": [38], "with": [41], "offline": [42], "training": [43, 109, 127], "processes.": [44], "However,": [45], "as": [46, 100], "increasing": [48], "demand": [49], "network": [51], "resilience,": [52], "further": [57], "expected": [58], "to": [59, 65], "be": [60, 66], "distributed,": [61], "and": [62, 95, 116, 163], "desirable": [64], "trained": [67], "an": [69], "online": [70, 108, 126], "fashion,": [72], "which": [73], "also": [75], "main": [77], "contributions": [78], "our": [80, 142, 168], "work.": [81], "To": [82], "solve": [83], "problem,": [85], "first": [87], "propose": [88], "a": [89, 101, 124, 177], "GRNN-based": [90, 144, 157], "method,": [94], "cast": [97], "self-supervised": [102], "problem.": [104], "Then,": [105], "is": [110, 129, 146, 159, 171], "achieved": [111], "via": [112], "gradient": [114], "computation,": [115], "inspired": [117], "by": [118, 148, 173], "(consensus-based)": [120], "optimization": [122], "idea,": [123], "optimizer": [128], "designed.": [130], "Furthermore,": [131], "local": [133, 161], "closed-loop": [134], "stability": [135], "system": [140], "under": [141], "proposed": [143, 169], "controller": [145, 158], "provided": [147], "assuming": [149], "that": [150], "nonlinear": [152], "activation": [153], "function": [154], "both": [160], "sector-bounded": [162], "slope-restricted.": [164], "The": [165], "effectiveness": [166], "method": [170], "illustrated": [172], "numerical": [174], "simulations": [175], "specifically": [178], "developed": [179], "simulator.": [180]}, "primary_topic": {"id": "https://openalex.org/T12794", "display_name": "Adaptive Dynamic Programming Control", "score": 0.7958999872207642, "subfield": {"id": "https://openalex.org/subfields/1703", "display_name": "Computational Theory and Mathematics"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T12794", "display_name": "Adaptive Dynamic Programming Control", "score": 0.7958999872207642, "subfield": {"id": "https://openalex.org/subfields/1703", "display_name": "Computational Theory and Mathematics"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10791", "display_name": "Advanced Control Systems Optimization", "score": 0.09830000251531601, "subfield": {"id": "https://openalex.org/subfields/2207", "display_name": "Control and Systems Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10249", "display_name": "Distributed Control Multi-Agent Systems", "score": 0.06369999796152115, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2504.06439", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2504.06439", "pdf_url": "https://arxiv.org/pdf/2504.06439", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W7108329353", "doi": "https://doi.org/10.6084/m9.figshare.30760940", "title": "Secure and Efficient Federated Learning for Predictive Modeling in Resource-Constrained Healthcare Systems", "publication_year": 2025, "publication_date": "2025-12-02", "type": "article", "language": null, "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Mirugwe, Alex", "orcid": "https://orcid.org/0000-0002-3194-4333", "observed_orcids": ["https://orcid.org/0000-0002-3194-4333"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Mirugwe, Alex", "raw_affiliation_strings": [], "raw_orcid": "https://orcid.org/0000-0002-3194-4333", "affiliations": []}, {"author_position": "last", "author": {"id": null, "display_name": "Nyirenda, Juwa", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Nyirenda, Juwa", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Predictive": [0], "modeling": [1], "in": [2, 11, 32, 86, 169, 188, 202, 216, 250], "healthcare": [3, 248], "holds": [4], "promise": [5], "for": [6, 93, 242], "improving": [7], "clinical": [8, 234], "outcomes,": [9], "but": [10], "many": [12], "low-resource": [13, 203], "settings,": [14], "data": [15, 49, 73, 104], "fragmentation,": [16], "privacy": [17, 39, 101, 184], "concerns,": [18], "and": [19, 43, 54, 77, 99, 153, 163, 173, 182, 237, 252], "infrastructural": [20], "limitations": [21], "hinder": [22], "centralized": [23, 149], "machine": [24], "learning": [25, 62], "approaches.": [26], "These": [27], "barriers": [28], "are": [29, 41], "especially": [30], "critical": [31], "human": [33], "immunodeficiency": [34], "virus": [35], "(HIV)": [36], "care,": [37], "where": [38], "concerns": [40], "heightened,": [42], "any": [44], "use": [45], "case": [46], "involving": [47], "patient-level": [48], "raises": [50], "significant": [51], "ethical,": [52], "regulatory,": [53], "confidentiality": [55], "concerns.": [56], "We": [57], "developed": [58], "a": [59, 148, 239], "privacy-preserving": [60, 213], "federated": [61, 91, 134], "(FL)": [63], "framework": [64, 89], "to": [65, 102, 115, 232], "predict": [66], "HIV": [67, 217], "viral": [68], "load": [69], "suppression": [70], "using": [71, 129, 196], "retrospective": [72], "from": [74], "50,000": [75], "patients": [76], "over": [78], "one": [79, 194], "million": [80], "visits": [81], "across": [82, 127, 246], "30": [83], "health": [84], "facilities": [85, 128], "Uganda.": [87], "The": [88, 133, 227], "utilizes": [90], "averaging": [92], "distributed": [94], "training,": [95], "secure": [96], "multiparty": [97], "aggregation,": [98], "differential": [100, 183], "ensure": [103], "confidentiality.": [105], "To": [106], "address": [107], "cross-site": [108], "heterogeneity,": [109], "we": [110], "integrated": [111], "domain-adversarial": [112], "neural": [113], "networks": [114], "promote": [116], "domain-invariant": [117], "feature": [118], "learning.": [119], "A": [120], "multilayer": [121], "perceptron": [122], "model": [123, 135], "was": [124, 191], "trained": [125], "collaboratively": [126], "only": [130], "local": [131], "data.": [132, 226], "achieved": [136], "an": [137], "area": [138], "under": [139], "the": [140, 221], "ROC": [141], "curve": [142], "(AUC)": [143], "of": [144, 223], "0.874,": [145], "nearly": [146], "matching": [147], "baseline": [150], "(AUC": [151], "0.881)": [152], "substantially": [154], "outperforming": [155], "site-specific": [156], "models": [157], "(average": [158], "AUC": [159], "0.758).": [160], "Sensitivity": [161], "(89.6%)": [162], "specificity": [164], "(66.8%)": [165], "demonstrate": [166], "strong": [167], "capability": [168], "identifying": [170], "both": [171], "suppressed": [172], "unsuppressed": [174], "cases.": [175], "Domain": [176], "adaptation": [177], "reduced": [178], "inter-facility": [179], "performance": [180, 215], "variability,": [181], "imposed": [185], "minimal": [186], "degradation": [187], "accuracy.": [189], "Training": [190], "completed": [192], "within": [193], "hour": [195], "modest": [197], "hardware,": [198], "which": [199], "supported": [200], "feasibility": [201], "settings.": [204], "Our": [205], "study": [206], "demonstrates": [207], "that": [208], "FL": [209], "can": [210], "deliver": [211], "robust,": [212], "predictive": [214], "care": [218], "without": [219], "requiring": [220], "centralization": [222], "sensitive": [224], "patient": [225], "proposed": [228], "architecture": [229], "is": [230], "adaptable": [231], "other": [233], "prediction": [235], "tasks": [236], "represents": [238], "practical": [240], "pathway": [241], "scaling": [243], "ethical": [244], "AI": [245], "decentralized": [247], "systems": [249], "low-": [251], "middle-income": [253], "countries.": [254]}, "primary_topic": {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 1.0, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 1.0, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T13702", "display_name": "Machine Learning in Healthcare", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11636", "display_name": "Artificial Intelligence in Healthcare and Education", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/2718", "display_name": "Health Informatics"}, "field": {"id": "https://openalex.org/fields/27", "display_name": "Medicine"}, "domain": {"id": "https://openalex.org/domains/4", "display_name": "Health Sciences"}}], "primary_location": {"id": "doi:10.6084/m9.figshare.30760940", "is_oa": true, "landing_page_url": "https://doi.org/10.6084/m9.figshare.30760940", "pdf_url": null, "source": {"id": "https://openalex.org/S4377196282", "display_name": "Figshare", "issn_l": null, "issn": null, "is_oa": false, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I4210132348", "host_organization_name": "Figshare (United Kingdom)", "host_organization_lineage": ["https://openalex.org/I4210132348"], "host_organization_lineage_names": [], "type": "repository"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": null, "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "JournalArticle"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W7163538789", "doi": "https://doi.org/10.64972/dea.2025.v3i1.68", "title": "Application of Parallelized Graph Neural Networks in Predicting Molecular Properties from Large-Scale Chemical Databases", "publication_year": 2025, "publication_date": "2025-01-13", "type": "article", "language": null, "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5137830422", "display_name": "Piotr Tomaszewski", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I11923345", "display_name": "Wroc\u0142aw University of Science and Technology", "ror": "https://ror.org/008fyn775", "country_code": "PL", "type": "education", "lineage": ["https://openalex.org/I11923345"]}], "countries": ["PL"], "is_corresponding": false, "raw_author_name": "Piotr Tomaszewski", "raw_affiliation_strings": ["Faculty of Electrical and Computer Engineering, Wrocaw University of Science and Technology, Wrocaw 50-370, Poland"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Faculty of Electrical and Computer Engineering, Wrocaw University of Science and Technology, Wrocaw 50-370, Poland", "institution_ids": ["https://openalex.org/I11923345"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5137869405", "display_name": "Adam Wojcik", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I864159182", "display_name": "University of Silesia in Katowice", "ror": "https://ror.org/0104rcc94", "country_code": "PL", "type": "education", "lineage": ["https://openalex.org/I864159182"]}], "countries": ["PL"], "is_corresponding": false, "raw_author_name": "Adam Wojcik", "raw_affiliation_strings": ["Faculty of Electrical Engineering, Automatics, Computer Science and Biomedical Engineering, University of Silesia in Katowice, Katowice 40-007, Poland"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Faculty of Electrical Engineering, Automatics, Computer Science and Biomedical Engineering, University of Silesia in Katowice, Katowice 40-007, Poland", "institution_ids": ["https://openalex.org/I864159182"]}]}], "abstract_inverted_index": {"A": [0], "parallel": [1, 30, 132], "GNN": [2, 45, 81, 133], "architecture": [3], "is": [4, 121], "proposed": [5], "to": [6, 38, 70, 136], "solve": [7], "the": [8, 71, 74, 79, 101], "problems": [9], "of": [10, 20, 73], "large": [11], "computation": [12], "and": [13, 28, 34, 66, 92, 112, 119, 126, 131], "poor": [14], "scalability": [15], "in": [16], "molecular": [17, 41, 140], "property": [18], "prediction": [19, 97, 141], "large-scale": [21, 139], "chemical": [22, 76], "databases.": [23], "This": [24], "study": [25], "uses": [26], "data-level": [27], "model-level": [29], "methods,": [31], "distributed": [32, 130], "training": [33, 46], "memory": [35, 90], "optimization": [36], "methods": [37, 134], "learn": [39], "graph-structured": [40], "data.": [42], "To": [43], "make": [44], "as": [47, 49], "efficient": [48], "possible": [50], "on": [51, 54], "heterogeneous": [52], "datasets": [53], "multi-node": [55], "GPU": [56], "clusters,": [57], "we": [58], "optimize": [59], "graph": [60], "batching,": [61], "adaptive": [62], "sampling,": [63], "workload": [64], "balancing,": [65], "communication-efficient": [67], "synchronization.": [68], "According": [69], "results": [72], "standard": [75], "benchmark": [77], "experiment,": [78], "parallelized": [80], "obtained": [82], "a": [83, 87, 95], "great": [84], "acceleration": [85], "ratio,": [86], "small": [88], "peak": [89], "usage,": [91], "still": [93], "had": [94], "high": [96], "accuracy": [98], "compared": [99], "with": [100], "basic": [102], "method.": [103], "The": [104], "comprehensive": [105], "analysis": [106], "analyzes": [107], "how": [108, 116], "using": [109], "better": [110], "partitioning": [111], "sampling": [113], "techniques": [114], "affects": [115], "many": [117], "conversations": [118], "what": [120], "needed.": [122], "In": [123], "computational": [124], "chemistry": [125], "materials": [127], "science,": [128], "these": [129], "seem": [135], "have": [137], "good": [138], "capabilities.": [142]}, "primary_topic": {"id": "https://openalex.org/T10211", "display_name": "Computational Drug Discovery Methods", "score": 0.6880000233650208, "subfield": {"id": "https://openalex.org/subfields/1703", "display_name": "Computational Theory and Mathematics"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10211", "display_name": "Computational Drug Discovery Methods", "score": 0.6880000233650208, "subfield": {"id": "https://openalex.org/subfields/1703", "display_name": "Computational Theory and Mathematics"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11948", "display_name": "Machine Learning in Materials Science", "score": 0.20960000157356262, "subfield": {"id": "https://openalex.org/subfields/2505", "display_name": "Materials Chemistry"}, "field": {"id": "https://openalex.org/fields/25", "display_name": "Materials Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10002", "display_name": "Advanced Chemical Physics Studies", "score": 0.035100001841783524, "subfield": {"id": "https://openalex.org/subfields/3107", "display_name": "Atomic and Molecular Physics, and Optics"}, "field": {"id": "https://openalex.org/fields/31", "display_name": "Physics and Astronomy"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.64972/dea.2025.v3i1.68", "is_oa": true, "landing_page_url": "https://doi.org/10.64972/dea.2025.v3i1.68", "pdf_url": "http://www.wpias.edu.pl/ojs/index.php/dea/article/download/68/78", "source": {"id": "https://openalex.org/S7407058643", "display_name": "Data Engineering and Applications", "issn_l": "3107-099X", "issn": ["3107-099X"], "is_oa": false, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": null, "host_organization_name": null, "host_organization_lineage": [], "host_organization_lineage_names": [], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Data Engineering and Applications", "raw_type": "journal-article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W7167484036", "doi": "https://doi.org/10.5281/zenodo.19152656", "title": "Edge Intelligence: Transforming Real-Time Data Processing in Smart Systems", "publication_year": 2025, "publication_date": "2025-01-13", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5140116988", "display_name": "Dirk R. Visser", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Dirk R. Visser", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5140103260", "display_name": "Camille A. Dupont", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Camille A. Dupont", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Edge": [0], "intelligence": [1, 7, 70, 185], "\u2014": [2, 18], "the": [3, 32], "deployment": [4, 186], "of": [5, 144, 153], "artificial": [6], "models": [8], "directly": [9], "at": [10], "network": [11], "edge": [12, 60, 69, 89, 124, 134, 184], "nodes": [13], "proximal": [14], "to": [15], "data": [16, 26], "sources": [17], "has": [19], "emerged": [20], "as": [21, 178], "acritical": [22], "paradigm": [23], "for": [24, 48, 77, 83, 92], "real-time": [25], "processing": [27], "in": [28, 187], "smart": [29, 105, 108, 119, 129, 188], "systems,": [30], "circumventing": [31], "latency,": [33], "bandwidth,": [34], "and": [35, 55, 87, 128, 150], "privacy": [36, 162], "constraints": [37], "imposed": [38], "bycentralised": [39], "cloud": [40], "architectures.": [41], "Despite": [42], "growing": [43], "industrial": [44], "adoption,": [45], "systematic": [46], "frameworks": [47], "co-optimising": [49], "model": [50, 79, 85], "accuracy,": [51], "inference": [52, 141], "latency,energy": [53], "consumption,": [54], "communication": [56, 170], "overhead": [57], "across": [58, 102], "heterogeneous": [59], "deployments": [61], "remain": [62], "underdeveloped.": [63], "This": [64], "study": [65], "presentsEdgeMind,": [66], "a": [67, 179], "unified": [68], "framework": [71], "integrating": [72], "neural": [73], "architecture": [74], "search": [75], "(NAS)": [76], "edge-optimised": [78], "design,": [80], "knowledgedistillation": [81], "(KD)": [82], "lightweight": [84], "deployment,": [86], "federated": [88], "learning": [90], "(FEL)": [91], "privacy-preserving": [93], "distributed": [94], "training": [95, 159], "acrossheterogeneous": [96], "IoT": [97], "nodes.": [98], "EdgeMind": [99, 139, 177], "was": [100], "evaluated": [101], "three": [103], "real-world": [104, 183], "system": [106], "deployments:": [107], "factory": [109], "anomaly": [110], "detection": [111, 132], "(48": [112], "edgenodes,": [113], "Siemens": [114], "S7": [115], "PLC": [116], "network,": [117], "Amsterdam),": [118], "city": [120], "traffic": [121], "monitoring": [122], "(64": [123], "cameras,": [125], "The": [126], "Hague),": [127], "grid": [130], "fault": [131], "(32substation": [133], "controllers,": [135], "Dutch": [136], "national": [137], "grid).": [138], "achieved": [140], "latency": [142], "reductions": [143, 148], "67.3%,": [145], "energy": [146], "consumption": [147], "of54.8%,": [149], "accuracy": [151], "improvements": [152], "8.4%": [154], "over": [155], "cloud-offloaded": [156], "baselines.": [157], "Federated": [158], "preserved": [160], "differential": [161], "(epsilon": [163], "=": [164], "1.2)": [165], "whileconverging": [166], "within": [167], "23.4%": [168], "fewer": [169], "rounds": [171], "than": [172], "FedAvg.": [173], "These": [174], "results": [175], "establish": [176], "scalable,": [180], "privacy-compliant": [181], "frameworkfor": [182], "systems.": [189]}, "primary_topic": {"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.9990000128746033, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.9990000128746033, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 9.999999747378752e-05, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 9.999999747378752e-05, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:ojs2.jisfc.com:article/7", "is_oa": true, "landing_page_url": "https://jisfc.com/index.php/jisfc/article/view/7", "pdf_url": "https://jisfc.com/index.php/jisfc/article/download/7/7", "source": {"id": "https://openalex.org/S7407079808", "display_name": "Journal of Intelligent Systems and Future Computing", "issn_l": "3117-4698", "issn": ["3117-4698"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": null, "host_organization_name": null, "host_organization_lineage": [], "host_organization_lineage_names": [], "type": "journal"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "3117-4698", "raw_type": "info:eu-repo/semantics/article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W7104515818", "doi": "https://doi.org/10.5281/zenodo.17564909", "title": "Scalable Deep Learning Models for WebScale Datasets", "publication_year": 2025, "publication_date": "2025-11-09", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Thorat, Ankur", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": true, "raw_author_name": "Thorat, Ankur", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"The": [0, 105, 127], "exponential": [1], "growth": [2], "of": [3, 20, 34, 111, 133, 141, 166, 169], "web-scale": [4, 45], "datasets": [5], "has": [6], "catalyzed": [7], "a": [8, 138, 155], "paradigm": [9], "shift": [10], "in": [11, 90], "deep": [12, 42, 135], "learningresearch": [13], "and": [14, 25, 29, 55, 82, 100, 120, 125, 160], "deployment.": [15], "Models": [16], "now": [17], "process": [18], "terabytes": [19], "multimodal": [21], "data,": [22], "ranging": [23], "fromtext": [24], "images": [26], "to": [27, 44, 49, 102], "video": [28], "code,": [30], "enabling": [31], "the": [32, 61, 109, 131], "emergence": [33], "foundation": [35], "models": [36], "with": [37], "broadgeneralization": [38], "abilities.": [39], "However,": [40], "scaling": [41], "learning": [43], "introducesunprecedented": [46], "challenges": [47], "related": [48], "computational": [50], "efficiency,": [51], "distributed": [52], "training,": [53], "memoryoptimization,": [54], "data": [56, 113, 145], "governance.": [57], "This": [58], "paper": [59], "explores": [60], "architectural,": [62], "algorithmic,": [63], "andinfrastructural": [64], "strategies": [65], "that": [66, 130], "make": [67], "scalability": [68], "possible,": [69], "focusing": [70], "on": [71, 137], "frameworks": [72], "such": [73], "asFully": [74], "Sharded": [75], "Data": [76], "Parallel": [77], "(FSDP),": [78], "ZeRO": [79], "redundancy": [80], "optimizer,": [81], "pipeline-parallel": [83], "modeltraining.": [84], "Furthermore,": [85], "it": [86], "analyzes": [87], "how": [88], "innovations": [89], "model": [91], "design\u2014such": [92], "as": [93], "sparsemixture-of-experts": [94], "(MoE)": [95], "networks,": [96], "attention": [97], "optimization": [98], "techniques,": [99], "quantization\u2014contribute": [101], "cost-effective": [103], "scalability.": [104], "research": [106], "also": [107], "highlights": [108], "importance": [110], "largescale": [112], "preprocessing": [114], "techniques": [115], "like": [116], "deduplication,": [117], "token": [118], "balancing,": [119], "noise": [121], "reductionto": [122], "ensure": [123], "quality": [124], "fairness.": [126], "study": [128], "concludes": [129], "future": [132], "scalable": [134], "learningdepends": [136], "holistic": [139], "integration": [140], "compute": [142], "optimization,": [143], "efficient": [144], "handling,": [146], "andresponsible": [147], "AI": [148, 163], "practices.": [149], "Through": [150], "this": [151], "synthesis,": [152], "we": [153], "provide": [154], "roadmap": [156], "for": [157], "building": [158], "robust,sustainable,": [159], "energy-efficient": [161], "large-scale": [162], "systems": [164], "capable": [165], "serving": [167], "billions": [168], "realworld": [170], "queries.": [171]}, "primary_topic": {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.32690000534057617, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.32690000534057617, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.26980000734329224, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.09939999878406525, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.5281/zenodo.17564909", "is_oa": true, "landing_page_url": "https://doi.org/10.5281/zenodo.17564909", "pdf_url": null, "source": {"id": "https://openalex.org/S4306400562", "display_name": "Zenodo (CERN European Organization for Nuclear Research)", "issn_l": null, "issn": null, "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I67311998", "host_organization_name": "European Organization for Nuclear Research", "host_organization_lineage": ["https://openalex.org/I67311998"], "host_organization_lineage_names": [], "type": "repository"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": null, "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "Preprint"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4417255548", "doi": "https://doi.org/10.48550/arxiv.2506.08132", "title": "Congestion-Aware Path Selection for Load Balancing in AI Clusters", "publication_year": 2025, "publication_date": "2025-06-09", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5120777244", "display_name": "Erfan Nosrati", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Nosrati, Erfan", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5081502954", "display_name": "Majid Ghaderi", "orcid": "https://orcid.org/0000-0002-3783-4346", "observed_orcids": ["https://orcid.org/0000-0002-3783-4346"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Ghaderi, Majid", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Fast": [0], "training": [1, 9, 22], "of": [2, 14, 16, 20, 28, 102, 167], "large": [3], "machine": [4], "learning": [5], "models": [6], "requires": [7, 122], "distributed": [8, 21], "on": [10, 25], "AI": [11, 93, 113], "clusters": [12], "consisting": [13], "thousands": [15], "GPUs.": [17], "The": [18], "efficiency": [19, 27], "crucially": [23], "depends": [24], "the": [26, 29, 34, 51, 98, 118, 134, 165, 191], "network": [30, 52, 86, 129], "interconnecting": [31], "GPUs": [32], "in": [33, 72, 92, 112], "cluster.": [35], "These": [36], "networks": [37], "are": [38, 76], "commonly": [39], "built": [40], "using": [41, 178], "RDMA": [42, 110], "following": [43], "a": [44, 104, 144, 155, 182], "Clos-like": [45], "datacenter": [46], "topology.": [47], "To": [48], "efficiently": [49], "utilize": [50], "bandwidth,": [53], "load": [54, 210], "balancing": [55, 211], "is": [56, 150], "employed": [57], "to": [58, 128, 143, 158, 170, 201, 207], "distribute": [59], "traffic": [60, 82, 111, 142], "across": [61], "multiple": [62], "redundant": [63], "paths.": [64], "While": [65], "there": [66], "exists": [67], "numerous": [68], "techniques": [69], "for": [70, 80, 109, 137], "load-balancing": [71, 106], "traditional": [73], "datacenters,": [74], "these": [75], "often": [77], "either": [78], "optimized": [79, 108], "TCP": [81], "or": [83, 126], "require": [84], "specialized": [85, 124], "hardware,": [87], "thus": [88], "limiting": [89], "their": [90], "utility": [91], "clusters.": [94, 114], "This": [95], "paper": [96], "presents": [97], "design": [99], "and": [100, 139, 162, 181, 193, 203], "evaluation": [101], "Hopper,": [103], "new": [105], "technique": [107], "Operating": [115], "entirely": [116], "at": [117], "host": [119], "level,": [120], "Hopper": [121, 177, 189], "no": [123], "hardware": [125], "modifications": [127], "switches.": [130], "It": [131], "continuously": [132], "monitors": [133], "current": [135], "path": [136, 147, 168], "congestion": [138, 149], "dynamically": [140], "switches": [141], "less": [145], "congested": [146], "when": [148], "detected.": [151], "Furthermore,": [152], "it": [153], "incorporates": [154], "lightweight": [156], "mechanism": [157], "identify": [159], "alternative": [160], "paths": [161], "carefully": [163], "controls": [164], "timing": [166], "switching": [169], "prevent": [171], "excessive": [172], "out-of-order": [173], "packets.": [174], "We": [175], "evaluated": [176], "ns-3": [179], "simulations": [180], "testbed": [183], "implementation.": [184], "Our": [185], "evaluations": [186], "show": [187], "that": [188], "reduces": [190], "average": [192], "99-percentile": [194], "tail": [195], "flow": [196], "completion": [197], "time": [198], "by": [199], "up": [200], "20%": [202], "14%,": [204], "respectively,": [205], "compared": [206], "state-of-the-art": [208], "host-based": [209], "techniques.": [212]}, "primary_topic": {"id": "https://openalex.org/T10138", "display_name": "Network Traffic and Congestion Control", "score": 0.9818999767303467, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10138", "display_name": "Network Traffic and Congestion Control", "score": 0.9818999767303467, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10829", "display_name": "Interconnection Networks and Systems", "score": 0.00419999985024333, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10714", "display_name": "Software-Defined Networks and 5G", "score": 0.003700000001117587, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2506.08132", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2506.08132", "pdf_url": "https://arxiv.org/pdf/2506.08132", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4415399006", "doi": "https://doi.org/10.1109/tcad.2025.3624135", "title": "AsyncGrid: An Intralayer and Interlayer Asynchronous Hybrid Parallelism System for Responsive Edge LLM Inference", "publication_year": 2025, "publication_date": "2025-10-21", "type": "article", "language": null, "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5035249009", "display_name": "Yi Xiong", "orcid": "https://orcid.org/0009-0002-7439-2505", "observed_orcids": ["https://orcid.org/0009-0002-7439-2505"]}, "institutions": [{"id": "https://openalex.org/I126520041", "display_name": "University of Science and Technology of China", "ror": "https://ror.org/04c4dkn09", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I126520041", "https://openalex.org/I19820366"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yi Xiong", "raw_affiliation_strings": ["School of Computer Science and Technology, University of Science and Technology of China, Hefei, China"], "raw_orcid": "https://orcid.org/0009-0002-7439-2505", "affiliations": [{"raw_affiliation_string": "School of Computer Science and Technology, University of Science and Technology of China, Hefei, China", "institution_ids": ["https://openalex.org/I126520041"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100421958", "display_name": "Rui Zhang", "orcid": "https://orcid.org/0000-0002-8132-6250", "observed_orcids": ["https://orcid.org/0000-0002-8132-6250", "https://orcid.org/0009-0000-9961-1108"]}, "institutions": [{"id": "https://openalex.org/I126520041", "display_name": "University of Science and Technology of China", "ror": "https://ror.org/04c4dkn09", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I126520041", "https://openalex.org/I19820366"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Rui Zhang", "raw_affiliation_strings": ["Suzhou Institute for Advanced Research, University of Science and Technology of China, Suzhou, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Suzhou Institute for Advanced Research, University of Science and Technology of China, Suzhou, China", "institution_ids": ["https://openalex.org/I126520041"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5117084700", "display_name": "Yulong Zu", "orcid": "https://orcid.org/0009-0003-2698-4343", "observed_orcids": ["https://orcid.org/0009-0003-2698-4343"]}, "institutions": [{"id": "https://openalex.org/I126520041", "display_name": "University of Science and Technology of China", "ror": "https://ror.org/04c4dkn09", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I126520041", "https://openalex.org/I19820366"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yulong Zu", "raw_affiliation_strings": ["Suzhou Institute for Advanced Research, University of Science and Technology of China, Suzhou, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Suzhou Institute for Advanced Research, University of Science and Technology of China, Suzhou, China", "institution_ids": ["https://openalex.org/I126520041"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5101644806", "display_name": "Weihong Liu", "orcid": "https://orcid.org/0000-0002-0308-0259", "observed_orcids": ["https://orcid.org/0000-0002-0308-0259", "https://orcid.org/0000-0002-7790-5700", "https://orcid.org/0009-0000-6782-4280"]}, "institutions": [{"id": "https://openalex.org/I126520041", "display_name": "University of Science and Technology of China", "ror": "https://ror.org/04c4dkn09", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I126520041", "https://openalex.org/I19820366"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Weihong Liu", "raw_affiliation_strings": ["School of Computer Science and Technology, University of Science and Technology of China, Hefei, China"], "raw_orcid": "https://orcid.org/0000-0002-0308-0259", "affiliations": [{"raw_affiliation_string": "School of Computer Science and Technology, University of Science and Technology of China, Hefei, China", "institution_ids": ["https://openalex.org/I126520041"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5022299432", "display_name": "Zongwei Zhu", "orcid": "https://orcid.org/0000-0003-3607-2631", "observed_orcids": ["https://orcid.org/0000-0003-3607-2631"]}, "institutions": [{"id": "https://openalex.org/I126520041", "display_name": "University of Science and Technology of China", "ror": "https://ror.org/04c4dkn09", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I126520041", "https://openalex.org/I19820366"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Zongwei Zhu", "raw_affiliation_strings": ["Suzhou Institute for Advanced Research, University of Science and Technology of China, Suzhou, China"], "raw_orcid": "https://orcid.org/0000-0003-3607-2631", "affiliations": [{"raw_affiliation_string": "Suzhou Institute for Advanced Research, University of Science and Technology of China, Suzhou, China", "institution_ids": ["https://openalex.org/I126520041"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5101916933", "display_name": "Jiawei Geng", "orcid": "https://orcid.org/0000-0001-7405-9820", "observed_orcids": ["https://orcid.org/0000-0001-7405-9820"]}, "institutions": [{"id": "https://openalex.org/I126520041", "display_name": "University of Science and Technology of China", "ror": "https://ror.org/04c4dkn09", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I126520041", "https://openalex.org/I19820366"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Jiawei Geng", "raw_affiliation_strings": ["School of Computer Science and Technology, University of Science and Technology of China, Hefei, China"], "raw_orcid": "https://orcid.org/0000-0001-7405-9820", "affiliations": [{"raw_affiliation_string": "School of Computer Science and Technology, University of Science and Technology of China, Hefei, China", "institution_ids": ["https://openalex.org/I126520041"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5103281432", "display_name": "Boyu Li", "orcid": "https://orcid.org/0009-0001-2796-2686", "observed_orcids": ["https://orcid.org/0009-0001-2796-2686"]}, "institutions": [{"id": "https://openalex.org/I126520041", "display_name": "University of Science and Technology of China", "ror": "https://ror.org/04c4dkn09", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I126520041", "https://openalex.org/I19820366"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Boyu Li", "raw_affiliation_strings": ["School of Computer Science and Technology, University of Science and Technology of China, Hefei, China"], "raw_orcid": "https://orcid.org/0009-0001-2796-2686", "affiliations": [{"raw_affiliation_string": "School of Computer Science and Technology, University of Science and Technology of China, Hefei, China", "institution_ids": ["https://openalex.org/I126520041"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5101273644", "display_name": "Qianyue Cao", "orcid": "https://orcid.org/0009-0001-2861-8211", "observed_orcids": ["https://orcid.org/0009-0001-2861-8211"]}, "institutions": [{"id": "https://openalex.org/I126520041", "display_name": "University of Science and Technology of China", "ror": "https://ror.org/04c4dkn09", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I126520041", "https://openalex.org/I19820366"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Qianyue Cao", "raw_affiliation_strings": ["School of Computer Science and Technology, University of Science and Technology of China, Hefei, China"], "raw_orcid": "https://orcid.org/0009-0001-2861-8211", "affiliations": [{"raw_affiliation_string": "School of Computer Science and Technology, University of Science and Technology of China, Hefei, China", "institution_ids": ["https://openalex.org/I126520041"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5077322091", "display_name": "Xuehai Zhou", "orcid": "https://orcid.org/0000-0002-8360-3143", "observed_orcids": ["https://orcid.org/0000-0002-8360-3143"]}, "institutions": [{"id": "https://openalex.org/I126520041", "display_name": "University of Science and Technology of China", "ror": "https://ror.org/04c4dkn09", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I126520041", "https://openalex.org/I19820366"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Xuehai Zhou", "raw_affiliation_strings": ["School of Computer Science and Technology, University of Science and Technology of China, Hefei, China"], "raw_orcid": "https://orcid.org/0000-0002-8360-3143", "affiliations": [{"raw_affiliation_string": "School of Computer Science and Technology, University of Science and Technology of China, Hefei, China", "institution_ids": ["https://openalex.org/I126520041"]}]}], "abstract_inverted_index": {"Edge": [0], "deployment": [1], "of": [2, 31, 40, 51, 74, 155, 226, 267], "large": [3], "language": [4], "models": [5, 246], "(LLMs)": [6], "is": [7, 83], "increasingly": [8], "attractive": [9], "due": [10, 119, 141], "to": [11, 95, 120, 142, 173, 181, 210, 260], "its": [12], "advantages": [13, 292], "in": [14, 25, 88, 160, 190, 299], "privacy,": [15], "customization,": [16], "and": [17, 35, 63, 162, 213, 274, 302], "availability.": [18], "However,": [19], "edge": [20, 65, 96, 241, 287], "environments": [21, 97], "face": [22], "significant": [23, 291], "challenges": [24], "reducing": [26], "Time-to-First-Token": [27], "(TTFT).": [28], "TTFT": [29, 219], "consists": [30], "(1)": [32, 245], "queuing": [33, 79], "delay": [34], "(2)": [36, 253], "prefill": [37, 61, 68, 107, 117, 130, 139, 175, 183], "latency,": [38], "both": [39, 300], "which": [41, 115, 168], "are": [42], "exacerbated": [43], "by": [44], "edge\u2011resource": [45], "constraints:": [46], "the": [47, 71, 153, 156, 223, 265], "substantial": [48, 297], "computational": [49], "demands": [50], "LLM": [52, 235, 242], "inference": [53, 236], "grow": [54], "superlinearly": [55], "with": [56, 197, 264], "prompt": [57], "length,": [58], "causing": [59], "high": [60, 138], "latency;": [62], "limited": [64], "resources": [66], "restrict": [67], "throughput,": [69], "preventing": [70], "timely": [72], "handling": [73], "incoming": [75], "requests,": [76], "thereby": [77], "exacerbating": [78], "delays.": [80], "Model": [81], "parallelism": [82, 101, 125, 200, 208], "a": [84, 202, 250], "commonly": [85], "used": [86], "solution": [87], "cloud-based": [89], "systems,": [90], "but": [91, 109, 136], "directly": [92], "applying": [93], "it": [94], "proves": [98], "ineffective.": [99], "Intra-layer": [100, 164], "(e.g.,": [102, 126], "tensor/sequence": [103], "parallelism)": [104, 128], "can": [105], "reduce": [106, 174], "latency": [108, 140, 176, 269], "suffers": [110], "from": [111], "frequent": [112], "global": [113, 179], "synchronization,": [114], "bottlenecks": [116], "throughput": [118, 131, 184, 272], "edge-limited": [121], "interconnection": [122], "bandwidth.": [123], "Inter-layer": [124], "pipeline": [127], "improves": [129], "via": [132], "fully": [133], "asynchronous": [134, 199], "execution": [135, 262], "retains": [137], "stage-wise": [143], "serialized": [144], "computation.": [145], "To": [146, 221], "address": [147], "this": [148, 150, 193, 227, 230], "dilemma,": [149], "paper": [151, 194, 231], "leverages": [152], "properties": [154], "causal": [157], "attention": [158], "mechanism": [159], "LLMs": [161], "proposes": [163], "Asynchronous": [165], "Parallelism": [166], "(IAP),": [167], "performs": [169], "intra-layer": [170, 191], "parallel": [171], "computations": [172], "while": [177, 270], "avoiding": [178], "synchronization": [180], "mitigate": [182], "bottlenecks.": [185], "Moreover,": [186], "considering": [187], "communication": [188, 278], "sensitivity": [189], "parallelism,": [192, 229], "integrate": [195], "IAP": [196], "inter-layer": [198], "into": [201], "unified": [203], "plan": [204], "space.": [205], "This": [206], "hybrid": [207, 228], "adapts": [209], "diverse": [211], "hardware": [212], "request": [214], "loads,": [215], "enabling": [216], "more": [217], "effective": [218], "optimization.": [220], "enable": [222], "end-to-end": [224], "implementation": [225], "propose": [232], "AsyncGrid,": [233], "an": [234, 255, 286], "system": [237], "tailored": [238], "for": [239], "responsive": [240], "inference.": [243], "AsyncGrid": [244], "runtime": [247], "overheads": [248], "through": [249], "performance": [251], "profiler,": [252], "employs": [254], "integer": [256], "programming": [257], "(IP)": [258], "formulation": [259], "optimize": [261], "plan,": [263], "objective": [266], "minimizing": [268], "meeting": [271], "requirements,": [273], "(3)": [275], "implements": [276], "fine-grained": [277], "optimization": [279], "during": [280], "runtime.": [281], "A": [282], "comprehensive": [283], "evaluation": [284], "on": [285], "testbed": [288], "demonstrates": [289], "AsyncGrid\u2019s": [290], "over": [293], "existing": [294], "methods,": [295], "achieving": [296], "improvements": [298], "homogeneous": [301], "heterogeneous": [303], "settings.": [304]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.5893999934196472, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.5893999934196472, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.18410000205039978, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.05130000039935112, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/tcad.2025.3624135", "is_oa": false, "landing_page_url": "https://doi.org/10.1109/tcad.2025.3624135", "pdf_url": null, "source": {"id": "https://openalex.org/S100835903", "display_name": "IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems", "issn_l": "0278-0070", "issn": ["0278-0070", "1937-4151"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-2", "norway-1"], "host_organization": "https://openalex.org/P4310319808", "host_organization_name": "Institute of Electrical and Electronics Engineers", "host_organization_lineage": ["https://openalex.org/P4310319808"], "host_organization_lineage_names": ["Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems", "raw_type": "journal-article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4416237593", "doi": "https://doi.org/10.48550/arxiv.2511.09741", "title": "TawPipe: Topology-Aware Weight Pipeline Parallelism for Accelerating Long-Context Large Models Training", "publication_year": 2025, "publication_date": "2025-11-12", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Wu, Houming", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Wu, Houming", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5089448955", "display_name": "Ling Chen", "orcid": "https://orcid.org/0000-0002-6490-3108", "observed_orcids": ["https://orcid.org/0000-0002-6490-3108", "https://orcid.org/0009-0000-8539-4683"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Chen, Ling", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Training": [0], "large": [1], "language": [2], "models": [3, 24, 155], "(LLMs)": [4], "is": [5], "fundamentally": [6], "constrained": [7], "by": [8, 22, 50], "limited": [9], "device": [10, 103], "memory": [11, 20], "and": [12, 64, 96, 110, 115, 162], "costly": [13], "inter-device": [14], "communication.": [15], "Although": [16], "pipeline": [17, 72], "parallelism": [18, 133], "alleviates": [19], "pressure": [21], "partitioning": [23], "across": [25], "devices,": [26], "it": [27], "incurs": [28], "activation": [29], "communication": [30, 83, 118, 138], "overhead": [31], "that": [32, 157], "scales": [33], "linearly": [34], "with": [35, 119, 153], "sequence": [36], "length,": [37], "limiting": [38], "efficiency": [39], "in": [40, 78, 129], "long-context": [41], "training.": [42], "Recent": [43], "weight-passing": [44], "approaches": [45], "(e.g.,": [46], "WeiPipe)": [47], "mitigate": [48], "this": [49], "transmitting": [51], "model": [52, 108], "weights": [53, 109], "instead": [54], "of": [55, 107], "activations,": [56], "but": [57], "suffer": [58], "from": [59], "redundant": [60, 113], "peer-to-peer": [61], "(P2P)": [62], "transfers": [63], "underutilized": [65], "intra-node": [66, 94], "bandwidth.": [67], "We": [68], "propose": [69], "TawPipe--topology-aware": [70], "weight": [71], "parallelism,": [73], "which": [74], "exploits": [75], "hierarchical": [76], "bandwidth": [77], "distributed": [79], "clusters": [80], "for": [81], "improved": [82], "efficiency.": [84], "TawPipe:": [85], "(i)": [86], "groups": [87], "devices": [88], "based": [89], "on": [90, 148], "topology": [91], "to": [92, 121, 150, 165], "optimize": [93], "collective": [95, 126], "inter-node": [97], "P2P": [98], "communication;": [99], "(ii)": [100], "assigns": [101], "each": [102], "a": [104], "fixed": [105], "shard": [106], "gradients,": [111], "avoiding": [112], "transfers;": [114], "(iii)": [116], "overlaps": [117], "computation": [120], "hide": [122], "latency.": [123], "Unlike": [124], "global": [125], "operations": [127], "used": [128], "fully": [130], "sharded": [131], "data": [132], "(FSDP),": [134], "TawPipe": [135, 158], "confines": [136], "most": [137], "within": [139], "node": [140], "boundaries,": [141], "significantly": [142], "reducing": [143], "cross-node": [144], "traffic.": [145], "Extensive": [146], "experiments": [147], "up": [149], "24": [151], "GPUs": [152], "LLaMA-style": [154], "show": [156], "achieves": [159], "superior": [160], "throughput": [161], "scalability": [163], "compared": [164], "state-of-the-art": [166], "baselines.": [167]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9639000296592712, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9639000296592712, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.004100000020116568, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.003700000001117587, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2511.09741", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2511.09741", "pdf_url": "https://arxiv.org/pdf/2511.09741", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4412862726", "doi": "https://doi.org/10.5120/ijca2025925445", "title": "System Design for AI Engineering: Adaptive Architectures for Real-World Scalable AI Applications", "publication_year": 2025, "publication_date": "2025-07-30", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5001473608", "display_name": "Abhishek Shukla", "orcid": "https://orcid.org/0009-0003-7348-2071", "observed_orcids": ["https://orcid.org/0009-0003-7348-2071"]}, "institutions": [{"id": "https://openalex.org/I70983195", "display_name": "Syracuse University", "ror": "https://ror.org/025r5qe02", "country_code": "US", "type": "education", "lineage": ["https://openalex.org/I70983195"]}], "countries": ["US"], "is_corresponding": true, "raw_author_name": "Abhishek Shukla", "raw_affiliation_strings": ["Syracuse University United States of America"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Syracuse University United States of America", "institution_ids": ["https://openalex.org/I70983195"]}]}], "abstract_inverted_index": {"The": [0], "rapid": [1], "advancement": [2], "of": [3], "Artificial": [4], "Intelligence": [5], "(AI)": [6], "necessitates": [7], "robust": [8], "system": [9, 64], "architectures": [10], "to": [11, 62], "ensure": [12], "scalability,": [13], "reliability,": [14], "and": [15, 43, 49, 59, 74, 81, 90, 101], "efficiency": [16], "across": [17], "diverse": [18], "applications.This": [19], "paper": [20], "proposes": [21], "a": [22, 107], "comprehensive": [23], "framework": [24, 104], "for": [25, 109], "designing": [26], "AI": [27, 100, 113], "engineering": [28], "systems,": [29, 71], "addressing": [30], "critical": [31], "components": [32], "such": [33, 84], "as": [34, 85, 106], "data": [35, 86], "pipelines,": [36], "computer": [37], "architectures,": [38, 58], "model": [39, 91], "serving,": [40], "distributed": [41], "training,": [42], "emerging": [44], "patterns": [45], "like": [46], "federated": [47], "learning": [48], "serverless": [50], "AI.We": [51], "introduce": [52], "novel": [53], "orchestration": [54], "techniques,": [55], "hybrid": [56], "cloud-edge": [57], "ethical": [60], "considerations": [61], "enhance": [63], "robustness.Through": [65], "detailed": [66], "case": [67], "studies": [68], "on": [69], "recommendation": [70], "autonomous": [72], "driving,": [73], "healthcare": [75], "diagnostics,": [76], "we": [77], "illustrate": [78], "practical": [79], "implementations": [80], "analyze": [82], "trade-offs.Challenges": [83], "privacy,": [87], "resource": [88], "optimization,": [89], "governance": [92], "are": [93], "explored,": [94], "with": [95], "future": [96], "directions": [97], "emphasizing": [98], "sustainable": [99], "quantum": [102], "computing.This": [103], "serves": [105], "blueprint": [108], "engineers": [110], "building": [111], "next-generation": [112], "systems.": [114]}, "primary_topic": {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.6650000214576721, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.6650000214576721, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T13904", "display_name": "Artificial Intelligence Applications", "score": 0.20080000162124634, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T12127", "display_name": "Software System Performance and Reliability", "score": 0.027499999850988388, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.5120/ijca2025925445", "is_oa": true, "landing_page_url": "https://doi.org/10.5120/ijca2025925445", "pdf_url": "https://doi.org/10.5120/ijca2025925445", "source": {"id": "https://openalex.org/S4210206007", "display_name": "International Journal of Computer Applications", "issn_l": "0975-8887", "issn": ["0975-8887"], "is_oa": true, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core"], "host_organization": "https://openalex.org/P4322545532", "host_organization_name": "Foundation of Computer Science", "host_organization_lineage": ["https://openalex.org/P4322545532"], "host_organization_lineage_names": ["Foundation of Computer Science"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "International Journal of Computer Applications", "raw_type": "journal-article"}, "cited_by_count": 1, "is_retracted": false}, {"id": "https://openalex.org/W4411713249", "doi": "https://doi.org/10.37745/ejcsit.2013/vol13n46124136", "title": "Federated Learning 3.0: A Self-Healing Framework for Cross-Cloud AI Training with Provable GDPR Compliance", "publication_year": 2025, "publication_date": "2025-05-15", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5118383132", "display_name": "Rahul Ganti", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Rahul Ganti", "raw_affiliation_strings": ["ERPA Inc. , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "ERPA Inc. , USA", "institution_ids": []}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5118659591", "display_name": "Vamsi Nellutla", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Vamsi Nellutla", "raw_affiliation_strings": ["Dallas Data Science Academy , USA"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Dallas Data Science Academy , USA", "institution_ids": []}]}], "abstract_inverted_index": {"Federated": [0], "Learning": [1], "3.0": [2], "represents": [3], "a": [4, 27, 141], "significant": [5], "advancement": [6], "in": [7, 92, 105], "privacy-preserving": [8], "artificial": [9], "intelligence": [10], "training": [11, 155], "across": [12], "diverse": [13], "regulatory": [14, 41, 146], "environments.": [15, 126], "This": [16, 127], "innovative": [17], "framework": [18, 101], "addresses": [19], "the": [20, 45, 131, 150], "fundamental": [21], "challenges": [22], "of": [23, 152], "cross-jurisdictional": [24], "compliance": [25, 49, 147], "through": [26, 69], "novel": [28], "self-healing": [29, 61], "architecture": [30], "built": [31], "on": [32], "blockchain": [33], "technology.": [34], "By": [35], "implementing": [36], "\"data": [37], "passports\"": [38], "that": [39], "track": [40], "context": [42], "and": [43, 51, 97, 114, 123], "permissions,": [44], "system": [46], "enables": [47], "real-time": [48], "monitoring": [50], "automatic": [52], "selective": [53], "forgetting": [54], "when": [55], "required": [56], "by": [57, 136], "regulations.": [58], "The": [59, 100], "architecture's": [60], "mechanism": [62], "maintains": [63], "model": [64], "performance": [65, 98], "after": [66], "data": [67], "removal": [68], "compensatory": [70], "re-weighting": [71], "techniques,": [72], "allowing": [73], "AI": [74], "systems": [75], "to": [76, 120], "operate": [77], "continuously": [78], "during": [79], "remediation": [80], "events.": [81], "Performance": [82], "benchmarks": [83], "demonstrate": [84], "substantial": [85], "improvements": [86], "over": [87], "traditional": [88], "federated": [89], "learning": [90], "approaches": [91], "erasure": [93], "speed,": [94], "audit": [95], "accuracy,": [96], "retention.": [99], "shows": [102], "particular": [103], "promise": [104], "healthcare": [106], "applications,": [107], "including": [108], "rare": [109], "disease": [110], "research,": [111], "pandemic": [112], "response,": [113], "personalized": [115], "medicine,": [116], "with": [117, 140], "potential": [118], "extensions": [119], "financial": [121], "services": [122], "edge": [124], "computing": [125], "approach": [128], "effectively": [129], "resolves": [130], "\"AI": [132], "amnesia\"": [133], "problem": [134], "identified": [135], "NIST,": [137], "providing": [138], "organizations": [139], "practical": [142], "solution": [143], "for": [144], "maintaining": [145], "while": [148], "leveraging": [149], "benefits": [151], "globally": [153], "distributed": [154], "data.": [156]}, "primary_topic": {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 1.0, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 1.0, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10270", "display_name": "Blockchain Technology Applications and Security", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11614", "display_name": "Cloud Data Security Solutions", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.37745/ejcsit.2013/vol13n46124136", "is_oa": true, "landing_page_url": "https://doi.org/10.37745/ejcsit.2013/vol13n46124136", "pdf_url": "https://eajournals.org/ejcsit/wp-content/uploads/sites/21/2025/06/Federated-Learning-3.0.pdf", "source": {"id": "https://openalex.org/S4210168180", "display_name": "European Journal of Computer Science and Information Technology", "issn_l": "2054-0957", "issn": ["2054-0957", "2054-0965"], "is_oa": false, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": null, "host_organization_name": null, "host_organization_lineage": [], "host_organization_lineage_names": [], "type": "journal"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "European Journal of Computer Science and Information Technology", "raw_type": "journal-article"}, "cited_by_count": 2, "is_retracted": false}, {"id": "https://openalex.org/W4416397567", "doi": "https://doi.org/10.1016/j.parco.2025.103164", "title": "LSHDP: Locally sharded heterogeneous data parallel for distributed deep learning", "publication_year": 2025, "publication_date": "2025-11-01", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5038788824", "display_name": "Motahhare Mirzaei", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I67009956", "display_name": "Iran University of Science and Technology", "ror": "https://ror.org/01jw2p796", "country_code": "IR", "type": "education", "lineage": ["https://openalex.org/I67009956"]}], "countries": ["IR"], "is_corresponding": false, "raw_author_name": "Motahhare Mirzaei", "raw_affiliation_strings": ["School of Computer Engineering, Iran University of Science and Technology, Tehran, Iran"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Computer Engineering, Iran University of Science and Technology, Tehran, Iran", "institution_ids": ["https://openalex.org/I67009956"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5037331610", "display_name": "Mehrdad Ashtiani", "orcid": "https://orcid.org/0000-0002-4574-1545", "observed_orcids": ["https://orcid.org/0000-0002-4574-1545"]}, "institutions": [{"id": "https://openalex.org/I67009956", "display_name": "Iran University of Science and Technology", "ror": "https://ror.org/01jw2p796", "country_code": "IR", "type": "education", "lineage": ["https://openalex.org/I67009956"]}], "countries": ["IR"], "is_corresponding": true, "raw_author_name": "Mehrdad Ashtiani", "raw_affiliation_strings": ["School of Computer Engineering, Iran University of Science and Technology, Tehran, Iran"], "raw_orcid": "https://orcid.org/0000-0002-4574-1545", "affiliations": [{"raw_affiliation_string": "School of Computer Engineering, Iran University of Science and Technology, Tehran, Iran", "institution_ids": ["https://openalex.org/I67009956"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5063102152", "display_name": "Mohammad Javad Pirhadi", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I67009956", "display_name": "Iran University of Science and Technology", "ror": "https://ror.org/01jw2p796", "country_code": "IR", "type": "education", "lineage": ["https://openalex.org/I67009956"]}], "countries": ["IR"], "is_corresponding": false, "raw_author_name": "Mohammad Javad Pirhadi", "raw_affiliation_strings": ["School of Computer Engineering, Iran University of Science and Technology, Tehran, Iran"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Computer Engineering, Iran University of Science and Technology, Tehran, Iran", "institution_ids": ["https://openalex.org/I67009956"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5086716685", "display_name": "Sauleh Eetemadi", "orcid": "https://orcid.org/0000-0003-1376-2023", "observed_orcids": ["https://orcid.org/0000-0003-1376-2023"]}, "institutions": [{"id": "https://openalex.org/I67009956", "display_name": "Iran University of Science and Technology", "ror": "https://ror.org/01jw2p796", "country_code": "IR", "type": "education", "lineage": ["https://openalex.org/I67009956"]}], "countries": ["IR"], "is_corresponding": false, "raw_author_name": "Sauleh Eetemadi", "raw_affiliation_strings": ["School of Computer Engineering, Iran University of Science and Technology, Tehran, Iran"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Computer Engineering, Iran University of Science and Technology, Tehran, Iran", "institution_ids": ["https://openalex.org/I67009956"]}]}], "abstract_inverted_index": null, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9053000211715698, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9053000211715698, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.03220000118017197, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.009999999776482582, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1016/j.parco.2025.103164", "is_oa": false, "landing_page_url": "https://doi.org/10.1016/j.parco.2025.103164", "pdf_url": null, "source": {"id": "https://openalex.org/S112708030", "display_name": "Parallel Computing", "issn_l": "0167-8191", "issn": ["0167-8191", "1872-7336"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-2", "norway-2"], "host_organization": "https://openalex.org/P4310320990", "host_organization_name": "Elsevier BV", "host_organization_lineage": ["https://openalex.org/P4310320990"], "host_organization_lineage_names": ["Elsevier BV"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Parallel Computing", "raw_type": "journal-article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4416038323", "doi": null, "title": "Mind Bubbles and Memory: Bounds on Scheduling Pipeline Parallelism with Rematerialization", "publication_year": 2025, "publication_date": "2025-09-01", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5119901262", "display_name": "Adrien Aguila--Multner", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I4405260729", "display_name": "TOPAL: Outils et Optimisations pour le Calcul Haute Performance et l'Apprentissage", "ror": "https://ror.org/046fmsw47", "country_code": "FR", "type": "facility", "lineage": ["https://openalex.org/I1326498283", "https://openalex.org/I4210131512", "https://openalex.org/I4405260729"]}], "countries": ["FR"], "is_corresponding": false, "raw_author_name": "Aguila--Multner, Adrien", "raw_affiliation_strings": ["TOPAL - Outils et Optimisations pour le Calcul Haute Performance et l'Apprentissage (200, avenue de la Vieille Tour 33405 Talence CEDEX - France)"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "TOPAL - Outils et Optimisations pour le Calcul Haute Performance et l'Apprentissage (200, avenue de la Vieille Tour 33405 Talence CEDEX - France)", "institution_ids": ["https://openalex.org/I4405260729"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5064515122", "display_name": "Olivier Beaumont", "orcid": "https://orcid.org/0000-0003-2741-6228", "observed_orcids": ["https://orcid.org/0000-0003-2741-6228"]}, "institutions": [{"id": "https://openalex.org/I4405260729", "display_name": "TOPAL: Outils et Optimisations pour le Calcul Haute Performance et l'Apprentissage", "ror": "https://ror.org/046fmsw47", "country_code": "FR", "type": "facility", "lineage": ["https://openalex.org/I1326498283", "https://openalex.org/I4210131512", "https://openalex.org/I4405260729"]}], "countries": ["FR"], "is_corresponding": false, "raw_author_name": "Beaumont, Olivier", "raw_affiliation_strings": ["TOPAL - Outils et Optimisations pour le Calcul Haute Performance et l'Apprentissage (200, avenue de la Vieille Tour 33405 Talence CEDEX - France)"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "TOPAL - Outils et Optimisations pour le Calcul Haute Performance et l'Apprentissage (200, avenue de la Vieille Tour 33405 Talence CEDEX - France)", "institution_ids": ["https://openalex.org/I4405260729"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5028817761", "display_name": "Lionel Eyraud\u2010Dubois", "orcid": "https://orcid.org/0000-0003-2475-3309", "observed_orcids": ["https://orcid.org/0000-0003-2475-3309"]}, "institutions": [{"id": "https://openalex.org/I4405260729", "display_name": "TOPAL: Outils et Optimisations pour le Calcul Haute Performance et l'Apprentissage", "ror": "https://ror.org/046fmsw47", "country_code": "FR", "type": "facility", "lineage": ["https://openalex.org/I1326498283", "https://openalex.org/I4210131512", "https://openalex.org/I4405260729"]}], "countries": ["FR"], "is_corresponding": false, "raw_author_name": "Eyraud-Dubois, Lionel", "raw_affiliation_strings": ["TOPAL - Outils et Optimisations pour le Calcul Haute Performance et l'Apprentissage (200, avenue de la Vieille Tour 33405 Talence CEDEX - France)"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "TOPAL - Outils et Optimisations pour le Calcul Haute Performance et l'Apprentissage (200, avenue de la Vieille Tour 33405 Talence CEDEX - France)", "institution_ids": ["https://openalex.org/I4405260729"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5034056852", "display_name": "Julia Gusak", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I1326498283", "display_name": "Institut national de recherche en sciences et technologies du num\u00e9rique", "ror": "https://ror.org/02kvxyf05", "country_code": "FR", "type": "government", "lineage": ["https://openalex.org/I1326498283"]}, {"id": "https://openalex.org/I4405260729", "display_name": "TOPAL: Outils et Optimisations pour le Calcul Haute Performance et l'Apprentissage", "ror": "https://ror.org/046fmsw47", "country_code": "FR", "type": "facility", "lineage": ["https://openalex.org/I1326498283", "https://openalex.org/I4210131512", "https://openalex.org/I4405260729"]}], "countries": ["FR"], "is_corresponding": false, "raw_author_name": "Gusak, Julia", "raw_affiliation_strings": ["Inria@SiliconValley - Inria@SiliconValley (United States)", "TOPAL - Outils et Optimisations pour le Calcul Haute Performance et l'Apprentissage (200, avenue de la Vieille Tour 33405 Talence CEDEX - France)"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Inria@SiliconValley - Inria@SiliconValley (United States)", "institution_ids": ["https://openalex.org/I1326498283"]}, {"raw_affiliation_string": "TOPAL - Outils et Optimisations pour le Calcul Haute Performance et l'Apprentissage (200, avenue de la Vieille Tour 33405 Talence CEDEX - France)", "institution_ids": ["https://openalex.org/I4405260729"]}]}], "abstract_inverted_index": {"Training": [0], "large": [1, 212], "neural": [2], "networks,": [3], "especially": [4], "Transformer-based": [5], "Large": [6], "Language": [7], "Models": [8], "(LLMs),": [9], "requires": [10], "massive": [11], "high-performance": [12], "computing": [13], "(HPC)": [14], "resources.": [15], "Within": [16], "each": [17, 124], "microbatch,": [18], "computations": [19], "follow": [20], "a": [21, 26, 31, 39, 80, 160, 183, 239], "strictly": [22], "sequential": [23, 46], "flow": [24], "through": [25], "stack": [27], "of": [28, 73, 75, 83, 155, 190], "transformer": [29], "blocks:": [30], "forward": [32], "pass": [33, 41], "to": [34, 42, 68, 106, 113], "compute": [35], "the": [36, 92, 104, 188, 229], "loss,": [37], "and": [38, 63, 99, 142, 176, 211, 235, 243], "backward": [40], "propagate": [43], "gradients.": [44], "This": [45, 77], "structure": [47], "limits": [48], "intrinsic": [49, 230], "parallelism.": [50], "To": [51], "improve": [52], "performance,": [53], "several": [54], "complementary": [55], "strategies": [56], "have": [57], "been": [58], "developed:": [59], "data,": [60], "tensor,": [61], "sequence,": [62], "pipeline": [64, 84, 105, 233, 245], "parallelism,": [65], "typically": [66], "combined": [67], "achieve": [69], "scalability": [70], "over": [71], "tens": [72], "thousands": [74], "GPUs.": [76], "paper": [78], "presents": [79], "formal": [81, 149], "analysis": [82, 180, 227], "parallelism": [85], "(PP)": [86], "for": [87, 164, 173, 202, 223, 241], "large-scale": [88], "training.": [89], "In": [90], "PP,": [91], "model": [93], "is": [94, 112], "partitioned": [95], "into": [96, 103], "multiple": [97, 131], "stages,": [98], "microbatches": [100], "are": [101, 200, 221], "injected": [102], "overlap": [107], "computation.": [108], "The": [109], "main": [110], "challenge": [111], "minimize": [114], "idle": [115, 145], "periods": [116], "(pipeline": [117], "bubbles)": [118], "while": [119, 217], "managing": [120], "memory": [121, 156, 184, 207, 213, 236, 250], "usage,": [122], "since": [123], "GPU": [125], "must": [126], "store": [127], "intermediate": [128, 224], "activations": [129, 191], "from": [130], "in-flight": [132], "microbatches.": [133], "Existing": [134], "scheduling": [135, 246], "algorithms": [136, 247], "such": [137], "as": [138], "GPIPE,": [139], "1F1B,": [140], "HANAYO,": [141], "MEGATRON": [143], "reduce": [144], "time": [146, 172], "but": [147], "lack": [148], "lower": [150, 168, 219], "bounds": [151, 169, 220], "or": [152], "explicit": [153], "modeling": [154], "constraints.": [157, 251], "We": [158], "develop": [159], "unified": [161], "analytical": [162], "approach": [163], "PP": [165], "scheduling,": [166], "deriving": [167], "on": [170], "completion": [171], "both": [174], "single-wave": [175], "multi-wave": [177], "regimes.": [178], "Our": [179, 226], "explicitly": [181], "incorporates": [182], "constraint": [185], "K,": [186], "denoting": [187], "number": [189], "that": [192], "can": [193], "be": [194], "stored": [195], "per": [196], "GPU.": [197], "Exact": [198], "results": [199], "provided": [201], "two": [203], "extreme": [204], "cases": [205], "(minimal": [206], "(K": [208, 214], "=": [209], "1)": [210], "\u2265": [215], "m)),": [216], "general": [218], "established": [222], "configurations.": [225], "highlights": [228], "coupling": [231], "between": [232], "utilization": [234], "footprint,": [237], "providing": [238], "foundation": [240], "evaluating": [242], "comparing": [244], "under": [248], "realistic": [249]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9970999956130981, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9970999956130981, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10715", "display_name": "Distributed and Parallel Computing Systems", "score": 0.00039999998989515007, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.0003000000142492354, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:HAL:hal-05353232v1", "is_oa": true, "landing_page_url": "https://inria.hal.science/hal-05353232", "pdf_url": null, "source": {"id": "https://openalex.org/S4306402512", "display_name": "HAL (Le Centre pour la Communication Scientifique Directe)", "issn_l": null, "issn": null, "is_oa": false, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I1294671590", "host_organization_name": "Centre National de la Recherche Scientifique", "host_organization_lineage": ["https://openalex.org/I1294671590"], "host_organization_lineage_names": [], "type": "repository"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": "2025", "raw_type": "info:eu-repo/semantics/preprint"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4413382354", "doi": "https://doi.org/10.1080/17538947.2025.2542913", "title": "A segment anything model-based geological remote sensing interpretation method with a distributed data-parallel deep learning framework", "publication_year": 2025, "publication_date": "2025-08-14", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5023386114", "display_name": "Xiaohui Huang", "orcid": "https://orcid.org/0000-0002-0394-2357", "observed_orcids": ["https://orcid.org/0000-0002-0394-2357"]}, "institutions": [{"id": "https://openalex.org/I3124059619", "display_name": "China University of Geosciences", "ror": "https://ror.org/04gcegc37", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I3124059619"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Xiaohui Huang", "raw_affiliation_strings": ["China University of Geosciences", "School of Computer Science, China University of Geosciences, Wuhan, People's Republic of China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "China University of Geosciences", "institution_ids": ["https://openalex.org/I3124059619"]}, {"raw_affiliation_string": "School of Computer Science, China University of Geosciences, Wuhan, People's Republic of China", "institution_ids": ["https://openalex.org/I3124059619"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5108411974", "display_name": "Ao Long", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I3124059619", "display_name": "China University of Geosciences", "ror": "https://ror.org/04gcegc37", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I3124059619"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Ao Long", "raw_affiliation_strings": ["China University of Geosciences", "School of Computer Science, China University of Geosciences, Wuhan, People's Republic of China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "China University of Geosciences", "institution_ids": ["https://openalex.org/I3124059619"]}, {"raw_affiliation_string": "School of Computer Science, China University of Geosciences, Wuhan, People's Republic of China", "institution_ids": ["https://openalex.org/I3124059619"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100750908", "display_name": "Wei Han", "orcid": "https://orcid.org/0000-0003-3882-1616", "observed_orcids": ["https://orcid.org/0000-0003-3882-1616"]}, "institutions": [{"id": "https://openalex.org/I3124059619", "display_name": "China University of Geosciences", "ror": "https://ror.org/04gcegc37", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I3124059619"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Wei Han", "raw_affiliation_strings": ["China University of Geosciences", "School of Computer Science, China University of Geosciences, Wuhan, People's Republic of China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "China University of Geosciences", "institution_ids": ["https://openalex.org/I3124059619"]}, {"raw_affiliation_string": "School of Computer Science, China University of Geosciences, Wuhan, People's Republic of China", "institution_ids": ["https://openalex.org/I3124059619"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5069537423", "display_name": "Yunliang Chen", "orcid": "https://orcid.org/0000-0001-9632-6192", "observed_orcids": ["https://orcid.org/0000-0001-9632-6192"]}, "institutions": [{"id": "https://openalex.org/I3124059619", "display_name": "China University of Geosciences", "ror": "https://ror.org/04gcegc37", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I3124059619"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yunliang Chen", "raw_affiliation_strings": ["China University of Geosciences", "School of Computer Science, China University of Geosciences, Wuhan, People's Republic of China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "China University of Geosciences", "institution_ids": ["https://openalex.org/I3124059619"]}, {"raw_affiliation_string": "School of Computer Science, China University of Geosciences, Wuhan, People's Republic of China", "institution_ids": ["https://openalex.org/I3124059619"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100770003", "display_name": "Geyong Min", "orcid": "https://orcid.org/0000-0003-1395-7314", "observed_orcids": ["https://orcid.org/0000-0003-1395-7314"]}, "institutions": [{"id": "https://openalex.org/I23923803", "display_name": "University of Exeter", "ror": "https://ror.org/03yghzc09", "country_code": "GB", "type": "education", "lineage": ["https://openalex.org/I23923803"]}], "countries": ["GB"], "is_corresponding": false, "raw_author_name": "Geyong Min", "raw_affiliation_strings": ["University of Exeter", "Department of Computer Science, University of Exeter, Exeter, UK"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "University of Exeter", "institution_ids": ["https://openalex.org/I23923803"]}, {"raw_affiliation_string": "Department of Computer Science, University of Exeter, Exeter, UK", "institution_ids": ["https://openalex.org/I23923803"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5101908495", "display_name": "Dongmei Yan", "orcid": "https://orcid.org/0009-0001-0271-8206", "observed_orcids": ["https://orcid.org/0009-0001-0271-8206"]}, "institutions": [{"id": "https://openalex.org/I19820366", "display_name": "Chinese Academy of Sciences", "ror": "https://ror.org/034t30j35", "country_code": "CN", "type": "government", "lineage": ["https://openalex.org/I19820366"]}, {"id": "https://openalex.org/I4210137199", "display_name": "Aerospace Information Research Institute", "ror": "https://ror.org/0419fj215", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I19820366", "https://openalex.org/I4210137199"]}, {"id": "https://openalex.org/I4210165038", "display_name": "University of Chinese Academy of Sciences", "ror": "https://ror.org/05qbk4x57", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I19820366", "https://openalex.org/I4210165038"]}, {"id": "https://openalex.org/I4405265946", "display_name": "International Research Center of Big Data for Sustainable Development Goals", "ror": "https://ror.org/027n0k361", "country_code": "CN", "type": "nonprofit", "lineage": ["https://openalex.org/I19820366", "https://openalex.org/I4405265946"]}], "countries": ["CN"], "is_corresponding": true, "raw_author_name": "Dongmei Yan", "raw_affiliation_strings": ["Chinese Academy of Sciences", "International Research Center of Big Data for Sustainable Development Goals", "University of Chinese Academy of Sciences", "Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, People's Republic of China", "Department of Big Earth Data Technologies & Products, International Research Center of Big Data for Sustainable Development Goals, Beijing, People's Republic of China", "Key Laboratory of Digital Earth Science, University of Chinese Academy of Sciences, Beijing, People's Republic of China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Chinese Academy of Sciences", "institution_ids": ["https://openalex.org/I19820366"]}, {"raw_affiliation_string": "International Research Center of Big Data for Sustainable Development Goals", "institution_ids": ["https://openalex.org/I4405265946"]}, {"raw_affiliation_string": "University of Chinese Academy of Sciences", "institution_ids": ["https://openalex.org/I4210165038"]}, {"raw_affiliation_string": "Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, People's Republic of China", "institution_ids": ["https://openalex.org/I19820366", "https://openalex.org/I4210137199"]}, {"raw_affiliation_string": "Department of Big Earth Data Technologies & Products, International Research Center of Big Data for Sustainable Development Goals, Beijing, People's Republic of China", "institution_ids": ["https://openalex.org/I4405265946"]}, {"raw_affiliation_string": "Key Laboratory of Digital Earth Science, University of Chinese Academy of Sciences, Beijing, People's Republic of China", "institution_ids": ["https://openalex.org/I4210165038"]}]}], "abstract_inverted_index": {"The": [0], "interpretation": [1, 47, 68], "of": [2, 12, 186], "remote": [3], "sensing": [4], "images": [5], "is": [6, 90], "pivotal": [7], "in": [8, 80, 169, 175], "extracting": [9, 121], "geological": [10, 27, 129], "elements": [11, 28], "interest.": [13], "Recent": [14], "studies": [15], "using": [16], "deep": [17], "learning": [18], "models": [19, 40], "often": [20], "fail": [21], "to": [22, 30, 142, 192], "provide": [23], "accurate": [24], "boundaries": [25, 53], "between": [26], "due": [29], "high": [31], "interclass": [32], "similarity": [33], "and": [34, 54, 64, 73, 125, 171, 189], "imbalanced": [35], "data": [36, 78], "distribution.": [37], "Furthermore,": [38], "these": [39], "are": [41], "computationally": [42], "intensive.": [43], "Inspired": [44], "by": [45, 123, 158], "expert": [46], "practices,": [48], "which": [49], "involve": [50], "first": [51], "delineating": [52], "then": [55], "identifying": [56], "semantics,": [57], "we": [58, 93], "leverage": [59], "the": [60, 95, 101, 105, 109, 113, 137, 144, 152, 155, 163], "vision": [61], "foundation": [62], "model": [63, 115], "propose": [65], "a": [66, 84, 132, 166, 172, 178], "distributed": [67, 71, 81], "framework": [69], "including": [70], "training": [72, 170, 194], "inference": [74, 110, 114, 176], "phases": [75], "based": [76], "on": [77, 177], "parallelism": [79], "architectures.": [82], "First,": [83], "conventional": [85], "semantic": [86, 133], "segmentation": [87], "model,": [88, 103], "DeepLabV3,": [89], "trained.": [91], "Subsequently,": [92], "integrate": [94], "Segment": [96], "Anything": [97], "Model": [98], "(SAM)": [99], "with": [100], "trained": [102], "completing": [104], "final": [106], "mapping": [107], "during": [108], "phase.": [111], "Specifically,": [112], "features": [116], "two": [117], "branches:": [118], "one": [119], "for": [120, 127, 146], "masks": [122], "SAM": [124], "another": [126], "categorizing": [128], "elements.": [130], "Then,": [131], "voting": [134], "module": [135], "combines": [136], "information": [138], "from": [139], "both": [140], "branches": [141], "determine": [143], "category": [145], "each": [147], "mask.": [148], "Experiments": [149], "demonstrate": [150], "that": [151], "method": [153, 164], "enhanced": [154], "performance": [156], "metric": [157], "more": [159], "than": [160], "2%.": [161], "Moreover,": [162], "achieves": [165], "3.85\u00d7": [167], "speedup": [168, 174], "3.95\u00d7": [173], "4-GPU": [179], "machine.": [180], "We": [181], "further": [182], "exploit": [183], "simulation": [184], "experiments": [185], "16": [187], "GPUs": [188], "achieve": [190], "up": [191], "15.1\u00d7": [193], "speedup.": [195]}, "primary_topic": {"id": "https://openalex.org/T10689", "display_name": "Remote-Sensing Image Classification", "score": 0.6151999831199646, "subfield": {"id": "https://openalex.org/subfields/2214", "display_name": "Media Technology"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10689", "display_name": "Remote-Sensing Image Classification", "score": 0.6151999831199646, "subfield": {"id": "https://openalex.org/subfields/2214", "display_name": "Media Technology"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.18250000476837158, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T12157", "display_name": "Geochemistry and Geologic Mapping", "score": 0.17440000176429749, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1080/17538947.2025.2542913", "is_oa": true, "landing_page_url": "https://doi.org/10.1080/17538947.2025.2542913", "pdf_url": null, "source": {"id": "https://openalex.org/S199162493", "display_name": "International Journal of Digital Earth", "issn_l": "1753-8947", "issn": ["1753-8947", "1753-8955"], "is_oa": true, "is_in_doaj": true, "is_core": true, "listed_in": ["cwts-core", "doaj", "erih-plus", "jufo-1", "norway-1"], "host_organization": "https://openalex.org/P4310320547", "host_organization_name": "Taylor & Francis", "host_organization_lineage": ["https://openalex.org/P4310320547", "https://openalex.org/P4310320449"], "host_organization_lineage_names": ["Taylor & Francis", "Informa"], "type": "journal"}, "license": "cc-by-nc", "license_id": "https://openalex.org/licenses/cc-by-nc", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "International Journal of Digital Earth", "raw_type": "journal-article"}, "cited_by_count": 1, "is_retracted": false}, {"id": "https://openalex.org/W4407638327", "doi": "https://doi.org/10.1109/jiot.2025.3542473", "title": "Resource Allocation for Task-Oriented Generative Artificial Intelligence in Internet of Things", "publication_year": 2025, "publication_date": "2025-02-17", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5101709392", "display_name": "Jie Feng", "orcid": "https://orcid.org/0000-0002-5474-3286", "observed_orcids": ["https://orcid.org/0000-0002-5474-3286"]}, "institutions": [{"id": "https://openalex.org/I149594827", "display_name": "Xidian University", "ror": "https://ror.org/05s92vm98", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I149594827"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Jie Feng", "raw_affiliation_strings": ["School of Telecommunication Engineering, Xidian University, Xi&#x2019;an, Shaanxi, China", "School of Telecomm. Engineering, Xidian University, Xi&#x2019;an, Shaanxi, China"], "raw_orcid": "https://orcid.org/0000-0002-5474-3286", "affiliations": [{"raw_affiliation_string": "School of Telecommunication Engineering, Xidian University, Xi&#x2019;an, Shaanxi, China", "institution_ids": ["https://openalex.org/I149594827"]}, {"raw_affiliation_string": "School of Telecomm. Engineering, Xidian University, Xi&#x2019;an, Shaanxi, China", "institution_ids": ["https://openalex.org/I149594827"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Xinqi Huang", "orcid": "https://orcid.org/0009-0008-3251-1821", "observed_orcids": ["https://orcid.org/0009-0008-3251-1821"]}, "institutions": [{"id": "https://openalex.org/I149594827", "display_name": "Xidian University", "ror": "https://ror.org/05s92vm98", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I149594827"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Xinqi Huang", "raw_affiliation_strings": ["School of Telecommunication Engineering, Xidian University, Xi&#x2019;an, Shaanxi, China", "School of Telecomm. Engineering, Xidian University, Xi&#x2019;an, Shaanxi, China"], "raw_orcid": "https://orcid.org/0009-0008-3251-1821", "affiliations": [{"raw_affiliation_string": "School of Telecommunication Engineering, Xidian University, Xi&#x2019;an, Shaanxi, China", "institution_ids": ["https://openalex.org/I149594827"]}, {"raw_affiliation_string": "School of Telecomm. Engineering, Xidian University, Xi&#x2019;an, Shaanxi, China", "institution_ids": ["https://openalex.org/I149594827"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100349496", "display_name": "Lei Liu", "orcid": "https://orcid.org/0000-0001-8173-0408", "observed_orcids": ["https://orcid.org/0000-0001-8173-0408"]}, "institutions": [{"id": "https://openalex.org/I149594827", "display_name": "Xidian University", "ror": "https://ror.org/05s92vm98", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I149594827"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Lei Liu", "raw_affiliation_strings": ["School of Telecommunication Engineering, Xidian University, Xi&#x2019;an, Shaanxi, China", "School of Telecomm. Engineering, Xidian University, Xi&#x2019;an, Shaanxi, China"], "raw_orcid": "https://orcid.org/0000-0001-8173-0408", "affiliations": [{"raw_affiliation_string": "School of Telecommunication Engineering, Xidian University, Xi&#x2019;an, Shaanxi, China", "institution_ids": ["https://openalex.org/I149594827"]}, {"raw_affiliation_string": "School of Telecomm. Engineering, Xidian University, Xi&#x2019;an, Shaanxi, China", "institution_ids": ["https://openalex.org/I149594827"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5087351635", "display_name": "Mengmeng Yang", "orcid": "https://orcid.org/0000-0002-8988-269X", "observed_orcids": ["https://orcid.org/0000-0002-8988-269X"]}, "institutions": [{"id": "https://openalex.org/I1292875679", "display_name": "Commonwealth Scientific and Industrial Research Organisation", "ror": "https://ror.org/03qn8fb07", "country_code": "AU", "type": "government", "lineage": ["https://openalex.org/I1292875679", "https://openalex.org/I2801453606", "https://openalex.org/I4387156119"]}, {"id": "https://openalex.org/I42894916", "display_name": "Data61", "ror": "https://ror.org/03q397159", "country_code": "AU", "type": "other", "lineage": ["https://openalex.org/I1292875679", "https://openalex.org/I2801453606", "https://openalex.org/I42894916", "https://openalex.org/I4387156119"]}], "countries": ["AU"], "is_corresponding": false, "raw_author_name": "Mengmeng Yang", "raw_affiliation_strings": ["Data61, Commonwealth Scientific and Industrial Research Organization, Melbourne, VIC, Australia", "Data61, Commonwealth Scientific and Industrial Research Organization, Australia"], "raw_orcid": "https://orcid.org/0000-0002-8988-269X", "affiliations": [{"raw_affiliation_string": "Data61, Commonwealth Scientific and Industrial Research Organization, Melbourne, VIC, Australia", "institution_ids": ["https://openalex.org/I42894916"]}, {"raw_affiliation_string": "Data61, Commonwealth Scientific and Industrial Research Organization, Australia", "institution_ids": ["https://openalex.org/I1292875679", "https://openalex.org/I42894916"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5065373445", "display_name": "Qingqi Pei", "orcid": "https://orcid.org/0000-0001-7601-5434", "observed_orcids": ["https://orcid.org/0000-0001-7601-5434", "https://orcid.org/0000-0001-7614-1422"]}, "institutions": [{"id": "https://openalex.org/I149594827", "display_name": "Xidian University", "ror": "https://ror.org/05s92vm98", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I149594827"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Qingqi Pei", "raw_affiliation_strings": ["School of Telecommunication Engineering, Xidian University, Xi&#x2019;an, Shaanxi, China", "School of Telecomm. Engineering, Xidian University, Xi&#x2019;an, Shaanxi, China"], "raw_orcid": "https://orcid.org/0000-0001-7601-5434", "affiliations": [{"raw_affiliation_string": "School of Telecommunication Engineering, Xidian University, Xi&#x2019;an, Shaanxi, China", "institution_ids": ["https://openalex.org/I149594827"]}, {"raw_affiliation_string": "School of Telecomm. Engineering, Xidian University, Xi&#x2019;an, Shaanxi, China", "institution_ids": ["https://openalex.org/I149594827"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5014395266", "display_name": "Yu Gang Shee", "orcid": "https://orcid.org/0000-0001-9833-3935", "observed_orcids": ["https://orcid.org/0000-0001-9833-3935"]}, "institutions": [{"id": "https://openalex.org/I4405256456", "display_name": "Universiti Tunku Abdul Rahman Sungai Long Campus", "ror": "https://ror.org/046b54093", "country_code": "MY", "type": "education", "lineage": ["https://openalex.org/I4405256456", "https://openalex.org/I931681460"]}, {"id": "https://openalex.org/I931681460", "display_name": "Universiti Tunku Abdul Rahman", "ror": "https://ror.org/050pq4m56", "country_code": "MY", "type": "education", "lineage": ["https://openalex.org/I931681460"]}], "countries": ["MY"], "is_corresponding": false, "raw_author_name": "Yu Gang Shee", "raw_affiliation_strings": ["Centre for Electronic Systems Design, Universiti Tunku Abdul Rahman, Kajang, Malaysia", "Centre for Electronic Systems Design, Universiti Tunku Abdul Rahman, Sungai Long Campus, Kajang, Selangor Darul Ehsan, Malaysia"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Centre for Electronic Systems Design, Universiti Tunku Abdul Rahman, Kajang, Malaysia", "institution_ids": ["https://openalex.org/I4405256456", "https://openalex.org/I931681460"]}, {"raw_affiliation_string": "Centre for Electronic Systems Design, Universiti Tunku Abdul Rahman, Sungai Long Campus, Kajang, Selangor Darul Ehsan, Malaysia", "institution_ids": ["https://openalex.org/I4405256456"]}]}], "abstract_inverted_index": {"The": [0, 88, 124, 175], "implementation": [1], "of": [2, 5, 16, 35], "the": [3, 10, 14, 32, 64, 85, 120, 160, 201], "Internet": [4], "Things": [6], "(IoT)": [7], "technology": [8], "has": [9], "potential": [11], "to": [12, 52, 118, 138], "unleash": [13], "capabilities": [15], "generative": [17], "artificial": [18], "intelligence": [19], "(GAI).": [20], "However,": [21], "integrating": [22], "GAI": [23, 48], "with": [24, 112], "IoT": [25, 50], "introduces": [26], "a": [27, 43, 57, 108, 128], "significant": [28], "challenge": [29], "in": [30, 49, 145], "managing": [31], "limited": [33], "resources": [34], "edge": [36, 206], "networks.": [37], "In": [38, 104], "this": [39, 54], "article,": [40], "we": [41, 106], "propose": [42], "resource": [44, 197], "optimization": [45], "framework": [46], "for": [47], "systems": [51], "address": [53], "issue,": [55], "leveraging": [56], "heterogeneous": [58, 74], "computing": [59, 75, 95], "framework.": [60], "We": [61], "focus": [62], "on": [63, 205], "system": [65], "utility": [66], "maximization": [67], "problem,": [68], "which": [69], "jointly": [70], "optimizes": [71], "transmit": [72], "power,": [73], "allocation,": [76], "CPU-cycle": [77, 90], "frequency,": [78, 80, 91, 93], "GPU-cycle": [79, 92], "and": [81, 94, 132, 142, 154, 171, 195], "task": [82, 121], "scheduling": [83, 122], "under": [84], "latency": [86], "constraint.": [87], "optimal": [89], "allocation": [96, 198], "are": [97], "obtained": [98], "by": [99, 148, 199], "employing": [100], "data": [101], "parallelism": [102], "analysis.": [103], "particular,": [105], "develop": [107], "hierarchical": [109, 129], "soft": [110], "actor-critic": [111], "an": [113, 133], "intrinsic": [114, 134], "curiosity": [115, 135], "(HSAC-IC)": [116], "algorithm": [117, 126, 162], "determine": [119], "strategy.": [123], "HSAC-IC": [125, 161], "utilizes": [127], "strategy": [130, 204], "structure": [131], "module": [136], "(ICM)": [137], "improve": [139], "learning": [140, 168], "efficiency": [141], "performance,": [143], "particularly": [144], "environments": [146], "characterized": [147], "sparse": [149], "rewards,": [150], "high-dimensional": [151], "action": [152], "spaces,": [153], "complex": [155], "tasks.": [156], "Our": [157], "simulations": [158], "benchmark": [159], "against": [163], "two": [164], "existing": [165], "deep": [166], "reinforcement": [167], "(DRL)": [169], "algorithms": [170], "three": [172], "reference": [173], "schemes.": [174], "results": [176], "illustrate": [177], "that": [178], "our": [179], "scheme": [180], "significantly": [181], "outperforms": [182], "these": [183], "alternatives,": [184], "ensuring": [185], "AIGC": [186], "user": [187], "service": [188, 192], "requirements,": [189], "while": [190], "minimizing": [191], "generation": [193], "costs,": [194], "optimizing": [196], "configuring": [200], "image": [202], "quality": [203], "servers.": [207]}, "primary_topic": {"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 1.0, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 1.0, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11409", "display_name": "Advanced Wireless Network Optimization", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/2208", "display_name": "Electrical and Electronic Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T13038", "display_name": "Internet of Things and AI", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/jiot.2025.3542473", "is_oa": false, "landing_page_url": "https://doi.org/10.1109/jiot.2025.3542473", "pdf_url": null, "source": {"id": "https://openalex.org/S2480266640", "display_name": "IEEE Internet of Things Journal", "issn_l": "2327-4662", "issn": ["2327-4662", "2372-2541"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-2", "norway-2"], "host_organization": "https://openalex.org/P4310319808", "host_organization_name": "Institute of Electrical and Electronics Engineers", "host_organization_lineage": ["https://openalex.org/P4310319808"], "host_organization_lineage_names": ["Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Internet of Things Journal", "raw_type": "journal-article"}, "cited_by_count": 12, "is_retracted": false}, {"id": "https://openalex.org/W4416071182", "doi": "https://doi.org/10.48550/arxiv.2506.00486", "title": "It Takes a Good Model to Train a Good Model: Generalized Gaussian Priors for Optimized LLMs", "publication_year": 2025, "publication_date": "2025-05-31", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5101435021", "display_name": "Jun Wu", "orcid": "https://orcid.org/0000-0002-8657-5475", "observed_orcids": ["https://orcid.org/0000-0002-8657-5475"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Wu, Jun", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": null, "display_name": "Huang, Patrick", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Huang, Patrick", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5103560215", "display_name": "Jiangtao Wen", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Wen, Jiangtao", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5101944724", "display_name": "Yuxing Han", "orcid": "https://orcid.org/0000-0003-2558-8966", "observed_orcids": ["https://orcid.org/0000-0003-2558-8966"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Han, Yuxing", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Despite": [0], "rapid": [1], "progress": [2], "in": [3, 33, 51, 98, 124], "large": [4], "language": [5], "models": [6, 109], "(LLMs),": [7], "the": [8], "statistical": [9, 126], "structure": [10], "of": [11], "their": [12], "weights,": [13], "activations,": [14], "and": [15, 23, 43, 70, 83, 86, 133], "gradients-and": [16], "its": [17], "implications": [18], "for": [19], "initialization,": [20], "training": [21, 78, 91], "dynamics,": [22], "efficiency-remains": [24], "largely": [25], "unexplored.": [26], "We": [27], "empirically": [28], "show": [29], "that": [30, 62, 80, 93, 114], "these": [31], "quantities": [32], "LLMs": [34], "are": [35, 56], "well": [36], "modeled": [37], "by": [38], "generalized": [39], "Gaussian": [40], "(GG)": [41], "distributions,": [42], "introduce": [44], "a": [45, 59, 75, 89], "unified,": [46], "end-to-end": [47], "optimization": [48, 123], "framework": [49], "grounded": [50], "this": [52, 128], "observation.": [53], "Our": [54], "contributions": [55], "threefold:": [57], "(1)": [58], "GG-based": [60], "initialization": [61], "aligns": [63], "with": [64, 110], "trained": [65], "model": [66], "statistics,": [67], "accelerating": [68], "convergence": [69], "improving": [71], "accuracy;": [72], "(2)": [73], "ACT,": [74], "progressive": [76], "activation-constrained": [77], "method": [79], "reduces": [81], "redundancy": [82], "propagation": [84], "overhead;": [85], "(3)": [87], "GCT,": [88], "gradient-constrained": [90], "algorithm": [92], "substantially": [94], "lowers": [95], "communication": [96, 112], "cost": [97], "distributed": [99], "training.": [100], "Experiments": [101], "across": [102], "diverse": [103], "architectures": [104], "demonstrate": [105], "consistently": [106], "smaller,": [107], "faster": [108], "minimal": [111], "overhead": [113], "match": [115], "or": [116], "surpass": [117], "standard": [118], "baselines.": [119], "By": [120], "anchoring": [121], "LLM": [122], "principled": [125], "modeling,": [127], "work": [129], "advances": [130], "efficient,": [131], "scalable,": [132], "hardware-aware": [134], "AI": [135], "systems.": [136]}, "primary_topic": {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.7695000171661377, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.7695000171661377, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.09489999711513519, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.04639999940991402, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2506.00486", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2506.00486", "pdf_url": "https://arxiv.org/pdf/2506.00486", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4415314754", "doi": "https://doi.org/10.48550/arxiv.2509.12213", "title": "Scaling Up Data Parallelism in Decentralized Deep Learning", "publication_year": 2025, "publication_date": "2025-08-31", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5101906407", "display_name": "Bing Xie", "orcid": "https://orcid.org/0000-0003-4500-0349", "observed_orcids": ["https://orcid.org/0000-0003-4500-0349"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Xie, Bing", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5051882926", "display_name": "Junqi Yin", "orcid": "https://orcid.org/0000-0003-3843-5520", "observed_orcids": ["https://orcid.org/0000-0003-3843-5520"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Yin, Junqi", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5039042766", "display_name": "Zhenyu Zhou", "orcid": "https://orcid.org/0000-0002-3344-4463", "observed_orcids": ["https://orcid.org/0000-0002-3344-4463", "https://orcid.org/0000-0001-7195-9313"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Zhou, Zhenyu", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5014538652", "display_name": "Sarp Oral", "orcid": "https://orcid.org/0000-0001-8745-7078", "observed_orcids": ["https://orcid.org/0000-0001-8745-7078"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Oral, Sarp", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5101916963", "display_name": "Feiyi Wang", "orcid": "https://orcid.org/0000-0002-0099-1559", "observed_orcids": ["https://orcid.org/0000-0002-0099-1559"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Wang, Feiyi", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Although": [0], "it": [1], "has": [2], "been": [3], "extensively": [4], "explored": [5], "in": [6, 27, 141, 191, 215], "theory,": [7], "decentralized": [8, 40, 45, 66, 110, 132, 150, 172, 183, 216], "learning": [9, 133, 151, 233], "is": [10, 134, 152], "not": [11], "yet": [12], "green-lighted": [13], "for": [14, 227, 239], "production": [15, 37], "use,": [16], "largely": [17], "due": [18], "to": [19, 61, 77, 107, 136, 155], "a": [20, 56, 74, 142, 171, 182], "lack": [21], "of": [22, 39, 87, 118, 131, 139, 149, 158, 244], "stability,": [23], "scalability,": [24], "and": [25, 65, 84, 94, 120, 186, 204, 219], "generality": [26, 121], "large": [28, 177, 201], "scale": [29, 178, 202, 243], "DNN": [30, 67, 179, 217], "training.": [31, 68], "To": [32, 51], "shed": [33], "light": [34], "on": [35, 98, 200, 241], "the": [36, 79, 85, 99, 116, 123, 137, 156, 166, 188, 210, 242], "use": [38, 192], "learning,": [41, 109], "this": [42, 52], "work": [43], "studies": [44], "data": [46, 111], "parallel": [47, 112], "training": [48, 95, 113, 124, 180, 195, 203, 237], "at": [49], "scale.": [50], "end,": [53], "we": [54, 72, 102, 168], "introduce": [55, 73], "benchmarking": [57, 75, 100], "framework,": [58], "namely": [59], "DBench,": [60, 71], "host": [62], "both": [63], "centralized": [64, 108, 232], "Building": [69], "upon": [70, 165], "methodology": [76], "uncover": [78], "correlations": [80], "between": [81], "model": [82, 129, 147, 162, 225], "accuracy": [83, 130, 148, 226], "variances": [86], "parameter": [88, 159], "tensors": [89, 160], "by": [90], "varying": [91], "communication": [92, 143, 189], "graphs": [93], "scales.": [96], "Based": [97], "results,": [101], "observe": [103, 205], "that,": [104], "(1)": [105], "Similar": [106], "also": [114], "presents": [115], "issues": [117], "scalability": [119], "when": [122, 236], "scales": [125], "up;": [126], "(2)": [127], "The": [128, 146], "correlated": [135], "number": [138], "connections": [140], "graph;": [144], "(3)": [145], "surprisingly": [153], "sensitive": [154], "variance": [157], "across": [161], "replicas.": [163], "Built": [164], "observations,": [167], "propose": [169], "Ada,": [170], "adaptive": [173], "approach": [174], "that": [175, 206], "performs": [176], "following": [181], "SGD": [184], "method": [185], "adapting": [187], "graph": [190], "dynamically": [193], "throughout": [194], "iterations.": [196], "We": [197], "apply": [198], "Ada": [199, 207], "can": [208], "obtain": [209], "best": [211], "convergence": [212], "rates": [213], "consistently": [214], "training,": [218], "delivers": [220], "equally": [221], "or": [222], "comparably": [223], "good": [224], "all": [228], "sample": [229], "applications": [230], "as": [231], "does,": [234], "even": [235], "ResNet50": [238], "ImageNet-1K": [240], "1008": [245], "GPUs.": [246]}, "primary_topic": {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.5468999743461609, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.5468999743461609, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.2517000138759613, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10249", "display_name": "Distributed Control Multi-Agent Systems", "score": 0.026100000366568565, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2509.12213", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2509.12213", "pdf_url": "https://arxiv.org/pdf/2509.12213", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4406094032", "doi": "https://doi.org/10.1109/tc.2025.3525604", "title": "Asynchronous Control Based Aggregation Transport Protocol for Distributed Deep Learning", "publication_year": 2025, "publication_date": "2025-01-06", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5101492539", "display_name": "Ye Jin", "orcid": "https://orcid.org/0000-0001-8087-6333", "observed_orcids": ["https://orcid.org/0000-0001-8087-6333"]}, "institutions": [{"id": "https://openalex.org/I150807315", "display_name": "Guangxi University", "ror": "https://ror.org/02c9qn167", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I150807315"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Jin Ye", "raw_affiliation_strings": ["School of Computer and Electronic Information, Guangxi University, Nanning, China"], "raw_orcid": "https://orcid.org/0000-0001-8087-6333", "affiliations": [{"raw_affiliation_string": "School of Computer and Electronic Information, Guangxi University, Nanning, China", "institution_ids": ["https://openalex.org/I150807315"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5057121591", "display_name": "Yajun Peng", "orcid": "https://orcid.org/0009-0001-6921-191X", "observed_orcids": ["https://orcid.org/0009-0001-6921-191X"]}, "institutions": [{"id": "https://openalex.org/I150807315", "display_name": "Guangxi University", "ror": "https://ror.org/02c9qn167", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I150807315"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yajun Peng", "raw_affiliation_strings": ["School of Computer and Electronic Information, Guangxi University, Nanning, China"], "raw_orcid": "https://orcid.org/0009-0001-6921-191X", "affiliations": [{"raw_affiliation_string": "School of Computer and Electronic Information, Guangxi University, Nanning, China", "institution_ids": ["https://openalex.org/I150807315"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5059077556", "display_name": "Yijun Li", "orcid": "https://orcid.org/0000-0003-4335-8742", "observed_orcids": ["https://orcid.org/0000-0003-4335-8742"]}, "institutions": [{"id": "https://openalex.org/I139660479", "display_name": "Central South University", "ror": "https://ror.org/00f1zfq44", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I139660479"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yijun Li", "raw_affiliation_strings": ["School of Computer Science and Engineering, Central South University, Changsha, China"], "raw_orcid": "https://orcid.org/0000-0003-4335-8742", "affiliations": [{"raw_affiliation_string": "School of Computer Science and Engineering, Central South University, Changsha, China", "institution_ids": ["https://openalex.org/I139660479"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5014213647", "display_name": "Zhaoyi Li", "orcid": "https://orcid.org/0000-0002-9677-2368", "observed_orcids": ["https://orcid.org/0000-0002-9677-2368", "https://orcid.org/0000-0001-6303-6041"]}, "institutions": [{"id": "https://openalex.org/I139660479", "display_name": "Central South University", "ror": "https://ror.org/00f1zfq44", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I139660479"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Zhaoyi Li", "raw_affiliation_strings": ["School of Computer Science and Engineering, Central South University, Changsha, China"], "raw_orcid": "https://orcid.org/0000-0002-9677-2368", "affiliations": [{"raw_affiliation_string": "School of Computer Science and Engineering, Central South University, Changsha, China", "institution_ids": ["https://openalex.org/I139660479"]}]}, {"author_position": "last", "author": {"id": null, "display_name": "Jiawei Huang", "orcid": "https://orcid.org/0000-0002-7578-4490", "observed_orcids": ["https://orcid.org/0000-0002-7578-4490"]}, "institutions": [{"id": "https://openalex.org/I139660479", "display_name": "Central South University", "ror": "https://ror.org/00f1zfq44", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I139660479"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Jiawei Huang", "raw_affiliation_strings": ["School of Computer Science and Engineering, Central South University, Changsha, China"], "raw_orcid": "https://orcid.org/0000-0002-7578-4490", "affiliations": [{"raw_affiliation_string": "School of Computer Science and Engineering, Central South University, Changsha, China", "institution_ids": ["https://openalex.org/I139660479"]}]}], "abstract_inverted_index": {"With": [0], "the": [1, 9, 25, 29, 48, 66, 73, 78, 107, 114, 119, 129, 135, 140, 155], "rapid": [2], "growth": [3], "scale": [4], "of": [5, 11, 81, 132, 139, 145], "dataset": [6], "and": [7, 60, 106, 134, 148, 162], "model,": [8], "training": [10, 71, 79, 165], "deep": [12], "neural": [13], "networks": [14], "(DNN)": [15], "tends": [16], "to": [17, 37, 52, 65, 103, 110, 117, 128, 160], "be": [18], "deployed": [19], "in": [20, 69, 166], "a": [21], "distributed": [22, 27], "manner.": [23], "In": [24], "large-scale": [26, 149], "training,": [28], "bottlenecks": [30], "have": [31], "gradually": [32], "moved": [33], "from": [34, 113], "computational": [35], "resources": [36], "communication": [38], "process.": [39], "Recent": [40], "researches": [41], "adopt": [42], "in-network": [43], "aggregation": [44, 50], "(INA)": [45], "that": [46, 152], "offloads": [47], "gradient": [49, 120], "process": [51], "programmable": [53], "switches,": [54], "thereby": [55], "reducing": [56], "network": [57], "traffic": [58], "amount": [59], "transmission": [61, 130], "latency.": [62], "Unfortunately,": [63], "due": [64], "bandwidth": [67], "competition": [68], "shared": [70], "clusters,": [72], "straggler": [74], "will": [75], "slow": [76], "down": [77], "efficiency": [80], "INA.": [82], "To": [83], "address": [84], "this": [85], "issue,": [86], "we": [87], "propose": [88], "an": [89], "Asynchronous": [90], "Control": [91], "based": [92], "Aggregation": [93], "Transport": [94], "Protocol": [95], "(AC-ATP),": [96], "which": [97], "makes": [98], "full": [99], "use": [100], "uncongested": [101], "links": [102], "transmit": [104], "gradients": [105, 112], "switch": [108], "memory": [109], "cache": [111], "fast": [115], "workers": [116], "accelerate": [118], "aggregation.": [121], "Meanwhile,": [122], "AC-ATP": [123, 153], "performs": [124], "congestion": [125], "control": [126], "according": [127], "progress": [131], "worker": [133], "remaining": [136], "completion": [137], "time": [138, 157], "job.": [141], "The": [142], "evaluation": [143], "results": [144], "real": [146], "testbed": [147], "simulations": [150], "show": [151], "reduces": [154], "aggregate": [156], "by": [158], "up": [159, 164], "68%": [161], "speeds": [163], "real-world": [167], "benchmark": [168], "models.": [169]}, "primary_topic": {"id": "https://openalex.org/T10138", "display_name": "Network Traffic and Congestion Control", "score": 0.49779999256134033, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10138", "display_name": "Network Traffic and Congestion Control", "score": 0.49779999256134033, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10714", "display_name": "Software-Defined Networks and 5G", "score": 0.11699999868869781, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.04809999838471413, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/tc.2025.3525604", "is_oa": false, "landing_page_url": "https://doi.org/10.1109/tc.2025.3525604", "pdf_url": null, "source": {"id": "https://openalex.org/S157670870", "display_name": "IEEE Transactions on Computers", "issn_l": "0018-9340", "issn": ["0018-9340", "0016-9340", "1557-9956", "2326-3814"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-3", "norway-2"], "host_organization": "https://openalex.org/P4310319808", "host_organization_name": "Institute of Electrical and Electronics Engineers", "host_organization_lineage": ["https://openalex.org/P4310319808"], "host_organization_lineage_names": ["Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Transactions on Computers", "raw_type": "journal-article"}, "cited_by_count": 3, "is_retracted": false}, {"id": "https://openalex.org/W4408357927", "doi": "https://doi.org/10.1109/tmc.2025.3550519", "title": "LVMScissor: Split and Schedule Large Vision Model Inference on Mobile Edges via Salp Swarm Algorithm", "publication_year": 2025, "publication_date": "2025-03-12", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5028848244", "display_name": "Yanting Liu", "orcid": "https://orcid.org/0000-0003-2265-8914", "observed_orcids": ["https://orcid.org/0000-0003-2265-8914"]}, "institutions": [{"id": "https://openalex.org/I14243506", "display_name": "Hong Kong Polytechnic University", "ror": "https://ror.org/0030zas98", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I14243506"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Yanting Liu", "raw_affiliation_strings": ["Department of Computing, The Hong Kong Polytechnic University, Hung Hom, Hong Kong"], "raw_orcid": "https://orcid.org/0000-0003-2265-8914", "affiliations": [{"raw_affiliation_string": "Department of Computing, The Hong Kong Polytechnic University, Hung Hom, Hong Kong", "institution_ids": ["https://openalex.org/I14243506"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5056801465", "display_name": "Rui Lu", "orcid": "https://orcid.org/0000-0002-2022-3872", "observed_orcids": ["https://orcid.org/0000-0002-2022-3872"]}, "institutions": [{"id": "https://openalex.org/I14243506", "display_name": "Hong Kong Polytechnic University", "ror": "https://ror.org/0030zas98", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I14243506"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Rui Lu", "raw_affiliation_strings": ["Department of Computing, The Hong Kong Polytechnic University, Hung Hom, Hong Kong"], "raw_orcid": "https://orcid.org/0000-0002-2022-3872", "affiliations": [{"raw_affiliation_string": "Department of Computing, The Hong Kong Polytechnic University, Hung Hom, Hong Kong", "institution_ids": ["https://openalex.org/I14243506"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5100411794", "display_name": "Dan Wang", "orcid": "https://orcid.org/0000-0002-0921-2726", "observed_orcids": ["https://orcid.org/0000-0002-0921-2726", "https://orcid.org/0000-0002-4727-897X"]}, "institutions": [{"id": "https://openalex.org/I14243506", "display_name": "Hong Kong Polytechnic University", "ror": "https://ror.org/0030zas98", "country_code": "HK", "type": "education", "lineage": ["https://openalex.org/I14243506"]}], "countries": ["HK"], "is_corresponding": false, "raw_author_name": "Dan Wang", "raw_affiliation_strings": ["Department of Computing, The Hong Kong Polytechnic University, Hung Hom, Hong Kong"], "raw_orcid": "https://orcid.org/0000-0002-0921-2726", "affiliations": [{"raw_affiliation_string": "Department of Computing, The Hong Kong Polytechnic University, Hung Hom, Hong Kong", "institution_ids": ["https://openalex.org/I14243506"]}]}], "abstract_inverted_index": {"In": [0], "the": [1, 4, 49, 64, 83, 118, 142], "Computer": [2], "Vision,": [3], "Large": [5], "Vision": [6, 11], "Models": [7], "(LVM)": [8], "based": [9, 136], "on": [10, 17, 27, 137], "Transformer": [12], "(ViT)": [13], "achieve": [14], "advanced": [15], "performance": [16], "general": [18], "complex": [19], "visual": [20], "tasks.": [21], "However,": [22], "deploying": [23], "these": [24], "resource-intensive": [25], "models": [26], "edge": [28, 42, 89, 173], "devices": [29, 90], "with": [30], "limited": [31], "computational": [32], "power": [33], "and": [34, 52, 85, 105, 152, 176], "memory": [35], "is": [36, 73, 168], "challenging,": [37], "especially": [38], "for": [39, 67, 75], "those": [40], "mobile": [41], "devices.": [43], "Existing": [44], "model": [45, 103, 150], "compression": [46], "works": [47], "downgrade": [48], "prediction": [50], "accuracy": [51], "fail": [53], "to": [54, 56, 148, 161], "adapt": [55], "dynamic": [57], "network": [58, 177], "bandwidth": [59, 86], "or": [60], "hardware": [61], "changes.": [62], "Besides,": [63], "split": [65, 97, 121, 153], "inference": [66, 98, 163], "typical": [68], "Deep": [69], "Neural": [70], "Networks": [71], "(DNN)": [72], "inefficient": [74], "LVM\u2019s": [76], "large": [77], "intermediate": [78], "result": [79], "size.": [80], "To": [81], "address": [82], "computation": [84], "limitation": [87], "of": [88, 91], "LVM,": [92], "we": [93, 130], "design": [94], "a": [95, 124, 133, 138], "new": [96], "acceleration": [99, 164], "LVMScissor": [100], "by": [101], "leveraging": [102], "parallelism": [104, 114, 151], "meta-heuristic": [106, 140], "algorithm.": [107], "We": [108], "first": [109], "implement": [110], "an": [111], "inter-layer": [112], "ViT": [113, 120], "strategy.": [115, 154], "After": [116], "modeling": [117], "parallelized": [119], "problem": [122], "into": [123], "Multi-task": [125], "three-processor": [126], "scheduling": [127, 134], "(MTS)": [128], "problem,": [129], "propose": [131], "LVM-MSSA,": [132], "algorithm": [135], "famous": [139], "algorithm,": [141], "Multi-objective": [143], "Salp": [144], "Swarm": [145], "Algorithm": [146], "(MSSA)": [147], "schedule": [149], "The": [155], "evaluation": [156], "results": [157], "show": [158], "that": [159], "compared": [160], "state-of-the-art": [162], "approaches,": [165], "our": [166], "solution": [167], "faster": [169], "from$1.6\\times$to$6.92\\times$under": [170], "variant": [171], "LVMs,": [172], "devices,": [174], "datasets,": [175], "traces.": [178]}, "primary_topic": {"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.8300999999046326, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.8300999999046326, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.15039999783039093, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.008100000210106373, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/tmc.2025.3550519", "is_oa": false, "landing_page_url": "https://doi.org/10.1109/tmc.2025.3550519", "pdf_url": null, "source": {"id": "https://openalex.org/S69141925", "display_name": "IEEE Transactions on Mobile Computing", "issn_l": "1536-1233", "issn": ["1536-1233", "1558-0660", "2161-9875"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-3", "norway-2"], "host_organization": "https://openalex.org/P4310320439", "host_organization_name": "IEEE Computer Society", "host_organization_lineage": ["https://openalex.org/P4310320439", "https://openalex.org/P4310319808"], "host_organization_lineage_names": ["IEEE Computer Society", "Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Transactions on Mobile Computing", "raw_type": "journal-article"}, "cited_by_count": 2, "is_retracted": false}, {"id": "https://openalex.org/W4415221398", "doi": "https://doi.org/10.1016/j.est.2025.118647", "title": "Design, transient simulation, and multi-objective optimization using artificial neural networks and genetic algorithms of a solar-assisted building energy system integrating energy storage: a case study of Xi'an, China", "publication_year": 2025, "publication_date": "2025-10-15", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5090857236", "display_name": "Gang Du", "orcid": "https://orcid.org/0000-0002-5143-2247", "observed_orcids": ["https://orcid.org/0000-0002-5143-2247"]}, "institutions": [{"id": "https://openalex.org/I4210162994", "display_name": "Xijing University", "ror": "https://ror.org/05xsjkb63", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I4210162994"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Gang Du", "raw_affiliation_strings": ["School of Computer Science, Xijing University, Xi'an, Shaanxi, 710123, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "School of Computer Science, Xijing University, Xi'an, Shaanxi, 710123, China", "institution_ids": ["https://openalex.org/I4210162994"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5101394865", "display_name": "Waqed H. Hassan", "orcid": "https://orcid.org/0000-0002-2351-2151", "observed_orcids": ["https://orcid.org/0000-0002-2351-2151"]}, "institutions": [{"id": "https://openalex.org/I4392021143", "display_name": "University of Warith Al-Anbiyaa", "ror": "https://ror.org/03ase0085", "country_code": null, "type": "education", "lineage": ["https://openalex.org/I4392021143"]}], "countries": [], "is_corresponding": true, "raw_author_name": "Waqed H. Hassan", "raw_affiliation_strings": ["Advanced Technical College, University of Warith Al-Anbiyaa, Karbala, Iraq"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Advanced Technical College, University of Warith Al-Anbiyaa, Karbala, Iraq", "institution_ids": ["https://openalex.org/I4392021143"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5037810969", "display_name": "Narinderjit Singh Sawaran Singh", "orcid": "https://orcid.org/0000-0003-3778-8711", "observed_orcids": ["https://orcid.org/0000-0003-3778-8711"]}, "institutions": [{"id": "https://openalex.org/I12649496", "display_name": "INTI International University", "ror": "https://ror.org/03fj82m46", "country_code": "MY", "type": "education", "lineage": ["https://openalex.org/I12649496"]}], "countries": ["MY"], "is_corresponding": false, "raw_author_name": "Narinderjit Singh Sawaran Singh", "raw_affiliation_strings": ["Faculty of Data Science and Information Technology, INTI International University, Persiaran Perdana BBN, Putra Nilai, Nilai, 71800, Malaysia"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Faculty of Data Science and Information Technology, INTI International University, Persiaran Perdana BBN, Putra Nilai, Nilai, 71800, Malaysia", "institution_ids": ["https://openalex.org/I12649496"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5005924357", "display_name": "Kamal Sharma", "orcid": "https://orcid.org/0000-0003-1863-2949", "observed_orcids": ["https://orcid.org/0000-0003-1863-2949"]}, "institutions": [{"id": "https://openalex.org/I82571370", "display_name": "GLA University", "ror": "https://ror.org/05fnxgv12", "country_code": "IN", "type": "education", "lineage": ["https://openalex.org/I82571370"]}], "countries": ["IN"], "is_corresponding": false, "raw_author_name": "Kamal Sharma", "raw_affiliation_strings": ["Department of Mechanical Engineering, Institute of Engineering and Technology, GLA University, Mathura, U.P., India"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Mechanical Engineering, Institute of Engineering and Technology, GLA University, Mathura, U.P., India", "institution_ids": ["https://openalex.org/I82571370"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5110311370", "display_name": "Abdullah Abed Hussein", "orcid": "https://orcid.org/0009-0009-5870-6309", "observed_orcids": ["https://orcid.org/0009-0009-5870-6309"]}, "institutions": [{"id": "https://openalex.org/I4405256833", "display_name": "University of Manara", "ror": "https://ror.org/04k20kq32", "country_code": "IQ", "type": "education", "lineage": ["https://openalex.org/I4405256833"]}], "countries": ["IQ"], "is_corresponding": false, "raw_author_name": "Abdullah Abed Hussein", "raw_affiliation_strings": ["Al-Manara College for Medical Sciences, Amarah, Maysan, Iraq"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Al-Manara College for Medical Sciences, Amarah, Maysan, Iraq", "institution_ids": ["https://openalex.org/I4405256833"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5019670010", "display_name": "Walid Aich", "orcid": "https://orcid.org/0000-0002-3373-2481", "observed_orcids": ["https://orcid.org/0000-0002-3373-2481", "https://orcid.org/0009-0008-1884-7919"]}, "institutions": [{"id": "https://openalex.org/I4210088963", "display_name": "University of Ha'il", "ror": "https://ror.org/013w98a82", "country_code": "SA", "type": "education", "lineage": ["https://openalex.org/I4210088963"]}], "countries": ["SA"], "is_corresponding": false, "raw_author_name": "Walid Aich", "raw_affiliation_strings": ["Department of Mechanical Engineering, College of Engineering, University of Ha'il, Ha'il City, 81451, Saudi Arabia"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Mechanical Engineering, College of Engineering, University of Ha'il, Ha'il City, 81451, Saudi Arabia", "institution_ids": ["https://openalex.org/I4210088963"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5058240476", "display_name": "Wajdi Rajhi", "orcid": "https://orcid.org/0000-0002-6963-3629", "observed_orcids": ["https://orcid.org/0000-0002-6963-3629"]}, "institutions": [{"id": "https://openalex.org/I4210088963", "display_name": "University of Ha'il", "ror": "https://ror.org/013w98a82", "country_code": "SA", "type": "education", "lineage": ["https://openalex.org/I4210088963"]}], "countries": ["SA"], "is_corresponding": false, "raw_author_name": "Wajdi Rajhi", "raw_affiliation_strings": ["Department of Mechanical Engineering, College of Engineering, University of Ha'il, Ha'il City, 81451, Saudi Arabia"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Mechanical Engineering, College of Engineering, University of Ha'il, Ha'il City, 81451, Saudi Arabia", "institution_ids": ["https://openalex.org/I4210088963"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5030798791", "display_name": "Saleh Alhumaid", "orcid": "https://orcid.org/0000-0002-6178-2240", "observed_orcids": ["https://orcid.org/0000-0002-6178-2240"]}, "institutions": [{"id": "https://openalex.org/I4210088963", "display_name": "University of Ha'il", "ror": "https://ror.org/013w98a82", "country_code": "SA", "type": "education", "lineage": ["https://openalex.org/I4210088963"]}], "countries": ["SA"], "is_corresponding": false, "raw_author_name": "Saleh Alhumaid", "raw_affiliation_strings": ["Department of Mechanical Engineering, College of Engineering, University of Ha'il, Ha'il City, 81451, Saudi Arabia"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Mechanical Engineering, College of Engineering, University of Ha'il, Ha'il City, 81451, Saudi Arabia", "institution_ids": ["https://openalex.org/I4210088963"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5117102020", "display_name": "Shaher Alshammrei", "orcid": "https://orcid.org/0000-0001-9838-920X", "observed_orcids": ["https://orcid.org/0000-0001-9838-920X"]}, "institutions": [{"id": "https://openalex.org/I4210088963", "display_name": "University of Ha'il", "ror": "https://ror.org/013w98a82", "country_code": "SA", "type": "education", "lineage": ["https://openalex.org/I4210088963"]}], "countries": ["SA"], "is_corresponding": false, "raw_author_name": "Shaher AlShammrei", "raw_affiliation_strings": ["Department of Mechanical Engineering, College of Engineering, University of Ha'il, Ha'il City, 81451, Saudi Arabia"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Department of Mechanical Engineering, College of Engineering, University of Ha'il, Ha'il City, 81451, Saudi Arabia", "institution_ids": ["https://openalex.org/I4210088963"]}]}], "abstract_inverted_index": null, "primary_topic": {"id": "https://openalex.org/T10121", "display_name": "Building Energy and Comfort Optimization", "score": 0.5360999703407288, "subfield": {"id": "https://openalex.org/subfields/2215", "display_name": "Building and Construction"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10121", "display_name": "Building Energy and Comfort Optimization", "score": 0.5360999703407288, "subfield": {"id": "https://openalex.org/subfields/2215", "display_name": "Building and Construction"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10905", "display_name": "Solar Thermal and Photovoltaic Systems", "score": 0.2754000127315521, "subfield": {"id": "https://openalex.org/subfields/2105", "display_name": "Renewable Energy, Sustainability and the Environment"}, "field": {"id": "https://openalex.org/fields/21", "display_name": "Energy"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T11185", "display_name": "Integrated Energy Systems Optimization", "score": 0.0828000009059906, "subfield": {"id": "https://openalex.org/subfields/2208", "display_name": "Electrical and Electronic Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1016/j.est.2025.118647", "is_oa": false, "landing_page_url": "https://doi.org/10.1016/j.est.2025.118647", "pdf_url": null, "source": {"id": "https://openalex.org/S2764865925", "display_name": "Journal of Energy Storage", "issn_l": "2352-152X", "issn": ["2352-152X", "2352-1538"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-2", "norway-1"], "host_organization": "https://openalex.org/P4310320990", "host_organization_name": "Elsevier BV", "host_organization_lineage": ["https://openalex.org/P4310320990"], "host_organization_lineage_names": ["Elsevier BV"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Journal of Energy Storage", "raw_type": "journal-article"}, "cited_by_count": 10, "is_retracted": false}, {"id": "https://openalex.org/W4416943185", "doi": "https://doi.org/10.1007/s11761-025-00479-2", "title": "Deep learning-driven tourist behavior prediction and intelligent recommendation system for smart tourism platforms", "publication_year": 2025, "publication_date": "2025-12-03", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5000275084", "display_name": "Li-an Tang", "orcid": "https://orcid.org/0009-0000-6054-4171", "observed_orcids": ["https://orcid.org/0009-0000-6054-4171"]}, "institutions": [{"id": "https://openalex.org/I4405279978", "display_name": "Taizhou Vocational College of Science and Technology", "ror": "https://ror.org/028rt6b30", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I4405279978"]}], "countries": ["CN"], "is_corresponding": true, "raw_author_name": "Lian Tang", "raw_affiliation_strings": ["Taizhou Vocational College of Science & Technology, Taizhou, 318000, Zhejiang, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Taizhou Vocational College of Science & Technology, Taizhou, 318000, Zhejiang, China", "institution_ids": ["https://openalex.org/I4405279978"]}]}], "abstract_inverted_index": null, "primary_topic": {"id": "https://openalex.org/T10203", "display_name": "Recommender Systems and Techniques", "score": 1.0, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10203", "display_name": "Recommender Systems and Techniques", "score": 1.0, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T12101", "display_name": "Advanced Bandit Algorithms Research", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/1803", "display_name": "Management Science and Operations Research"}, "field": {"id": "https://openalex.org/fields/18", "display_name": "Decision Sciences"}, "domain": {"id": "https://openalex.org/domains/2", "display_name": "Social Sciences"}}, {"id": "https://openalex.org/T10286", "display_name": "Information Retrieval and Search Behavior", "score": 0.0, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1007/s11761-025-00479-2", "is_oa": false, "landing_page_url": "https://doi.org/10.1007/s11761-025-00479-2", "pdf_url": null, "source": {"id": "https://openalex.org/S4210170441", "display_name": "Service Oriented Computing and Applications", "issn_l": "1863-2386", "issn": ["1863-2386", "1863-2394"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-1", "norway-1"], "host_organization": "https://openalex.org/P4310319900", "host_organization_name": "Springer Science+Business Media", "host_organization_lineage": ["https://openalex.org/P4310319900", "https://openalex.org/P4310319965"], "host_organization_lineage_names": ["Springer Science+Business Media", "Springer Nature"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Service Oriented Computing and Applications", "raw_type": "journal-article"}, "cited_by_count": 3, "is_retracted": false}, {"id": "https://openalex.org/W4414973959", "doi": "https://doi.org/10.48550/arxiv.2510.05186", "title": "OptPipe: Memory- and Scheduling-Optimized Pipeline Parallelism for LLM Training", "publication_year": 2025, "publication_date": "2025-10-06", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5079361316", "display_name": "Hongpei Li", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Li, Hongpei", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100399317", "display_name": "Han Zhang", "orcid": "https://orcid.org/0000-0002-5608-9185", "observed_orcids": ["https://orcid.org/0000-0002-5608-9185"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Zhang, Han", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": null, "display_name": "Liu, Huikang", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Liu, Huikang", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": null, "display_name": "Ge, Dongdong", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Ge, Dongdong", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5038083443", "display_name": "Yuhan Ye", "orcid": "https://orcid.org/0009-0007-2495-7586", "observed_orcids": ["https://orcid.org/0009-0007-2495-7586", "https://orcid.org/0009-0001-3312-0434"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Ye, Yinyu", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Pipeline": [0], "parallelism": [1], "(PP)": [2], "has": [3], "become": [4], "a": [5, 57, 93, 139], "standard": [6], "technique": [7], "for": [8, 100, 136], "scaling": [9], "large": [10], "language": [11], "model": [12, 111, 150], "(LLM)": [13], "training": [14, 191], "across": [15], "multiple": [16], "devices.": [17], "However,": [18], "despite": [19], "recent": [20], "progress": [21], "in": [22, 138, 186], "reducing": [23], "memory": [24, 81, 101, 123, 135, 166, 183, 197], "consumption": [25], "through": [26], "activation": [27, 103], "offloading,": [28], "existing": [29, 128], "approaches": [30, 133], "remain": [31], "largely": [32], "heuristic": [33], "and": [34, 44, 83, 105, 152, 165, 185], "coarse-grained,": [35], "often": [36], "overlooking": [37], "the": [38, 52, 78, 145, 180, 190], "fine-grained": [39, 113], "trade-offs": [40], "between": [41, 80], "memory,": [42], "computation,": [43], "scheduling": [45, 54, 84, 91], "latency.": [46], "In": [47, 168], "this": [48, 110], "work,": [49], "we": [50, 89, 142, 170], "revisit": [51], "pipeline": [53, 106, 117, 173], "problem": [55, 96], "from": [56], "principled": [58], "optimization": [59, 95], "perspective.": [60], "We": [61], "observe": [62], "that": [63, 97, 115, 158], "prevailing": [64], "strategies": [65], "either": [66], "rely": [67], "on": [68], "static": [69], "rules": [70], "or": [71], "aggressively": [72], "offload": [73], "activations": [74], "without": [75], "fully": [76], "leveraging": [77], "interaction": [79], "constraints": [82], "efficiency.": [85], "To": [86], "address": [87], "this,": [88], "formulate": [90], "as": [92], "constrained": [94], "jointly": [98], "accounts": [99], "capacity,": [102], "reuse,": [104], "bubble": [107], "minimization.": [108], "Solving": [109], "yields": [112], "schedules": [114], "reduce": [116, 171], "bubbles": [118], "while": [119], "adhering": [120], "to": [121, 149, 177], "strict": [122], "budgets.": [124, 198], "Our": [125], "approach": [126], "complements": [127], "offloading": [129], "techniques:": [130], "whereas": [131], "prior": [132], "trade": [134], "time": [137, 174], "fixed": [140], "pattern,": [141], "dynamically": [143], "optimize": [144], "tradeoff": [146], "with": [147], "respect": [148], "structure": [151], "hardware": [153], "configuration.": [154], "Experimental": [155], "results": [156], "demonstrate": [157], "our": [159], "method": [160], "consistently": [161], "improves": [162], "both": [163], "throughput": [164], "utilization.": [167], "particular,": [169], "idle": [172], "by": [175], "up": [176], "50%": [178], "under": [179], "same": [181], "per-device": [182], "limit,": [184], "some": [187], "cases,": [188], "enable": [189], "of": [192], "larger": [193], "models": [194], "within": [195], "limited": [196]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9922000169754028, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9922000169754028, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.0012000000569969416, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.0005000000237487257, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2510.05186", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2510.05186", "pdf_url": "https://arxiv.org/pdf/2510.05186", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4415930308", "doi": "https://doi.org/10.3390/electronics14214333", "title": "FADP-GT: A Frequency-Adaptive and Dual-Pooling Graph Transformer Model for Device Placement in Model Parallelism", "publication_year": 2025, "publication_date": "2025-11-05", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Hao Shu", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I119868032", "display_name": "Shanxi Agricultural University", "ror": "https://ror.org/05e9f5362", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I119868032"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Hao Shu", "raw_affiliation_strings": ["College of Software, Shanxi Agricultural University, Jinzhong 030801, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "College of Software, Shanxi Agricultural University, Jinzhong 030801, China", "institution_ids": ["https://openalex.org/I119868032"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5041739881", "display_name": "Wangli Hao", "orcid": "https://orcid.org/0000-0003-4674-0330", "observed_orcids": ["https://orcid.org/0000-0003-4674-0330"]}, "institutions": [{"id": "https://openalex.org/I119868032", "display_name": "Shanxi Agricultural University", "ror": "https://ror.org/05e9f5362", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I119868032"]}], "countries": ["CN"], "is_corresponding": true, "raw_author_name": "Wangli Hao", "raw_affiliation_strings": ["College of Software, Shanxi Agricultural University, Jinzhong 030801, China"], "raw_orcid": "https://orcid.org/0000-0003-4674-0330", "affiliations": [{"raw_affiliation_string": "College of Software, Shanxi Agricultural University, Jinzhong 030801, China", "institution_ids": ["https://openalex.org/I119868032"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5002157315", "display_name": "Meng Han", "orcid": "https://orcid.org/0000-0003-2868-6246", "observed_orcids": ["https://orcid.org/0000-0003-2868-6246"]}, "institutions": [{"id": "https://openalex.org/I119868032", "display_name": "Shanxi Agricultural University", "ror": "https://ror.org/05e9f5362", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I119868032"]}, {"id": "https://openalex.org/I50760025", "display_name": "Hangzhou Dianzi University", "ror": "https://ror.org/0576gt767", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I50760025"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Meng Han", "raw_affiliation_strings": ["Hangzhou Economic Development Zone, Hangzhou Dianzi University, Hangzhou 310018, China", "School of Information Science and Engineering, Shanxi Agricultural University, Jinzhong 030801, China"], "raw_orcid": "https://orcid.org/0000-0003-2868-6246", "affiliations": [{"raw_affiliation_string": "Hangzhou Economic Development Zone, Hangzhou Dianzi University, Hangzhou 310018, China", "institution_ids": ["https://openalex.org/I50760025"]}, {"raw_affiliation_string": "School of Information Science and Engineering, Shanxi Agricultural University, Jinzhong 030801, China", "institution_ids": ["https://openalex.org/I119868032"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5101951918", "display_name": "Fuzhong Li", "orcid": "https://orcid.org/0000-0003-0558-9988", "observed_orcids": ["https://orcid.org/0000-0003-0558-9988", "https://orcid.org/0009-0007-9189-8379", "https://orcid.org/0000-0001-9695-8309"]}, "institutions": [{"id": "https://openalex.org/I119868032", "display_name": "Shanxi Agricultural University", "ror": "https://ror.org/05e9f5362", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I119868032"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Fuzhong Li", "raw_affiliation_strings": ["College of Software, Shanxi Agricultural University, Jinzhong 030801, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "College of Software, Shanxi Agricultural University, Jinzhong 030801, China", "institution_ids": ["https://openalex.org/I119868032"]}]}], "abstract_inverted_index": {"The": [0, 18, 76, 188], "increasing": [1], "scale": [2], "and": [3, 69, 94, 109, 134, 148, 207], "complexity": [4], "of": [5, 20, 27, 92, 168, 182], "graph-structured": [6], "data": [7], "necessitate": [8], "efficient": [9, 169], "parallel": [10], "training": [11], "strategies": [12, 22], "for": [13, 47, 202], "graph": [14, 28, 126, 158], "neural": [15], "networks": [16], "(GNNs).": [17], "effectiveness": [19], "these": [21], "hinges": [23], "on": [24], "the": [25, 84, 90, 102, 111, 119, 122, 139, 166, 178, 198, 208], "quality": [26], "feature": [29, 45, 146, 159, 175], "representation.": [30], "To": [31], "this": [32], "end,": [33], "we": [34], "propose": [35, 51], "a": [36, 52, 63, 70], "Frequency-Adaptive": [37, 53, 64], "Dual-Pooling": [38, 54, 71], "Graph": [39, 55, 65], "Transformer": [40, 56], "(FADP-GT)": [41, 57], "model": [42, 163], "to": [43, 87, 105, 152, 213], "enhance": [44], "learning": [46], "computational": [48, 186], "graphs.": [49, 187], "We": [50], "model,": [58], "which": [59, 142], "incorporates": [60], "two": [61], "modules:": [62], "Attention": [66], "(FAGA)": [67], "module": [68, 78, 124], "Feature": [72], "Refinement": [73], "(DPFR)": [74], "module.": [75], "FAGA": [77], "adaptively": [79], "filters": [80], "frequency": [81], "components": [82], "in": [83, 97, 114, 185], "spectral": [85], "domain": [86], "dynamically": [88], "adjust": [89], "contribution": [91], "high-": [93], "low-frequency": [95], "information": [96, 108], "attention": [98], "computation,": [99], "thereby": [100], "enhancing": [101], "model\u2019s": [103], "ability": [104], "capture": [106], "structural": [107], "mitigating": [110], "over-smoothing": [112], "problem": [113], "multi-layer": [115], "network": [116], "propagation.": [117], "On": [118], "other": [120], "hand,": [121], "DPFR": [123], "refines": [125], "features": [127], "through": [128], "dual-pooling": [129], "operations\u2014Global": [130], "Average": [131], "Pooling": [132, 137], "(GAP)": [133], "Global": [135], "Max": [136], "(GMP)\u2014along": [138], "node": [140, 183], "dimension,": [141], "captures": [143], "both": [144], "global": [145], "distributions": [147], "salient": [149], "local": [150], "patterns": [151], "enrich": [153], "multi-scale": [154], "representations.": [155], "By": [156], "improving": [157], "representation,": [160], "our": [161], "FADP-GT": [162, 193], "indirectly": [164], "supports": [165], "development": [167], "device": [170, 203], "placement": [171, 204], "strategies,": [172], "as": [173], "enhanced": [174], "extraction": [176], "enables": [177], "more": [179], "accurate": [180], "modeling": [181], "dependencies": [184], "experimental": [189], "results": [190], "demonstrate": [191], "that": [192], "outperforms": [194], "existing": [195], "methods,": [196], "reducing": [197], "average": [199], "computation": [200], "time": [201, 210], "by": [205, 211], "96.52%": [206], "execution": [209], "9.06%": [212], "26.48%.": [214]}, "primary_topic": {"id": "https://openalex.org/T11273", "display_name": "Advanced Graph Neural Networks", "score": 0.8988999724388123, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T11273", "display_name": "Advanced Graph Neural Networks", "score": 0.8988999724388123, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T12292", "display_name": "Graph Theory and Algorithms", "score": 0.0877000018954277, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.0024999999441206455, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.3390/electronics14214333", "is_oa": true, "landing_page_url": "https://doi.org/10.3390/electronics14214333", "pdf_url": "https://www.mdpi.com/2079-9292/14/21/4333/pdf?version=1762331096", "source": {"id": "https://openalex.org/S4210202905", "display_name": "Electronics", "issn_l": "2079-9292", "issn": ["2079-9292"], "is_oa": true, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "norway-1"], "host_organization": "https://openalex.org/P4310310987", "host_organization_name": "Multidisciplinary Digital Publishing Institute", "host_organization_lineage": ["https://openalex.org/P4310310987"], "host_organization_lineage_names": ["Multidisciplinary Digital Publishing Institute"], "type": "journal"}, "license": "cc-by", "license_id": "https://openalex.org/licenses/cc-by", "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "Electronics", "raw_type": "journal-article"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4407690036", "doi": "https://doi.org/10.48550/arxiv.2502.11034", "title": "AdaGC: Enhancing LLM Pretraining Stability via Adaptive Gradient Clipping", "publication_year": 2025, "publication_date": "2025-02-16", "type": "preprint", "language": "en", "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5049245418", "display_name": "Guoxia Wang", "orcid": "https://orcid.org/0000-0002-3145-3892", "observed_orcids": ["https://orcid.org/0000-0002-3145-3892", "https://orcid.org/0009-0004-3915-5613", "https://orcid.org/0000-0002-7872-7977", "https://orcid.org/0009-0003-5494-7937"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Wang, Guoxia", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100424071", "display_name": "Shuai Li", "orcid": "https://orcid.org/0000-0001-9884-8250", "observed_orcids": ["https://orcid.org/0000-0001-9884-8250"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Li, Shuai", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5027220244", "display_name": "Congliang Chen", "orcid": "https://orcid.org/0000-0002-9795-4200", "observed_orcids": ["https://orcid.org/0000-0002-9795-4200"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Chen, Congliang", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": null, "display_name": "Zeng, Jinle", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Zeng, Jinle", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5060015858", "display_name": "Jiabin Yang", "orcid": "https://orcid.org/0000-0002-6182-990X", "observed_orcids": ["https://orcid.org/0000-0002-6182-990X"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Yang, Jiabin", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5034102301", "display_name": "Dingli Yu", "orcid": "https://orcid.org/0000-0003-3177-8942", "observed_orcids": ["https://orcid.org/0000-0003-3177-8942"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Yu, Dianhai", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100307788", "display_name": "Yanjun Ma", "orcid": "https://orcid.org/0000-0002-0521-8434", "observed_orcids": ["https://orcid.org/0000-0002-0521-8434"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Ma, Yanjun", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": "https://openalex.org/A5100333326", "display_name": "Shen Li", "orcid": "https://orcid.org/0000-0002-9482-3045", "observed_orcids": ["https://orcid.org/0000-0002-9482-3045", "https://orcid.org/0000-0001-5659-3464"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Shen, Li", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Loss": [0], "spikes": [1, 23, 34, 46, 72], "remain": [2], "a": [3, 50, 92, 113], "persistent": [4], "obstacle": [5], "in": [6, 31, 138], "large-scale": [7], "language": [8], "model": [9], "pretraining.": [10], "While": [11], "previous": [12], "research": [13], "has": [14], "attempted": [15], "to": [16, 112, 135, 162], "identify": [17], "the": [18, 39, 68], "root": [19], "cause": [20], "of": [21, 41, 52, 67, 118], "loss": [22, 45], "by": [24, 38, 107, 173], "investigating": [25], "individual": [26], "factors,": [27], "we": [28, 90], "observe": [29], "that,": [30], "practice,": [32], "such": [33, 105, 185], "are": [35], "typically": [36], "triggered": [37], "confluence": [40], "heterogeneous": [42], "factors.": [43], "Empirically,": [44], "may": [47], "arise": [48], "from": [49], "combination": [51], "data": [53], "outliers,": [54], "hardware": [55], "or": [56], "transient": [57], "computational": [58], "faults,": [59], "numerical": [60], "precision": [61], "issues,": [62], "and": [63, 84, 130, 148, 167, 176, 188, 195], "hyperparameter": [64], "settings.": [65], "Regardless": [66], "underlying": [69], "cause,": [70], "these": [71], "manifest": [73], "as": [74, 78, 186], "unstable": [75], "optimizer": [76], "updates,": [77], "abnormal": [79], "gradients": [80], "contaminate": [81], "both": [82], "first-": [83], "second-moment": [85], "states.": [86], "In": [87], "this": [88], "paper,": [89], "propose": [91], "principled": [93], "gradient-centric": [94], "remedy:": [95], "AdaGC,": [96], "an": [97], "adaptive": [98], "per-tensor": [99], "gradient": [100, 109], "clipping": [101], "scheme": [102], "that": [103, 152], "mitigates": [104], "contamination": [106], "bounding": [108], "norms": [110], "relative": [111], "tensor-wise": [114], "exponential": [115], "moving": [116], "average": [117, 193], "their": [119], "historical": [120], "clipped": [121], "values.": [122], "AdaGC": [123, 153, 180], "is": [124, 201], "optimizer-agnostic,": [125], "introduces": [126], "negligible": [127], "memory": [128], "overhead,": [129], "reduces": [131], "communication": [132], "costs": [133], "compared": [134], "GlobalGC,": [136], "particularly": [137], "hybrid-parallel": [139], "distributed": [140], "training.": [141], "Experiments": [142], "on": [143], "Llama-2": [144], "7B,": [145], "Mixtral": [146], "8x1B,": [147], "ERNIE": [149], "10B-A1.4B": [150], "demonstrate": [151], "robustly": [154], "eliminates": [155], "training": [156], "instabilities,": [157], "consistently": [158, 190], "reducing": [159], "spike": [160, 197], "scores": [161], "zero": [163, 196], "for": [164], "all": [165], "models": [166], "improving": [168], "downstream": [169], "accuracy": [170, 194], "over": [171], "GlobalGC": [172], "1.32%,": [174], "1.27%,": [175], "2.48%,": [177], "respectively.": [178], "Furthermore,": [179], "seamlessly": [181], "integrates": [182], "with": [183], "optimizers": [184], "Muon": [187], "Lion,": [189], "yielding": [191], "higher": [192], "scores.": [198], "The": [199], "code": [200], "available": [202], "at": [203], "https://github.com/PaddlePaddle/PaddleFleet": [204], "(see": [205], "Research/AdaGC).": [206]}, "primary_topic": {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.625, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.625, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.2759000062942505, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T12535", "display_name": "Machine Learning and Data Classification", "score": 0.024700000882148743, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2502.11034", "is_oa": true, "landing_page_url": "https://arxiv.org/abs/2502.11034", "pdf_url": "https://arxiv.org/pdf/2502.11034", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4415885801", "doi": "https://doi.org/10.48550/arxiv.2510.15830", "title": "SNOO: Step-K Nesterov Outer Optimizer - The Surprising Effectiveness of Nesterov Momentum Applied to Pseudo-Gradients", "publication_year": 2025, "publication_date": "2025-10-17", "type": "preprint", "language": null, "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5120252856", "display_name": "Dominik Kallusky", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Kallusky, Dominik", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": null, "display_name": "Rao, Vinay", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Rao, Vinay", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5120252857", "display_name": "Vishal Nandavanam", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Nandavanam, Vishal", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}, {"author_position": "last", "author": {"id": null, "display_name": "Shi, Hao-Jun Michael", "orcid": null, "observed_orcids": []}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Shi, Hao-Jun Michael", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"The": [0], "rapid": [1], "development": [2], "of": [3, 22, 33, 132, 144, 156, 176], "large": [4], "language": [5], "models": [6], "(LLMs)": [7], "has": [8], "driven": [9], "the": [10, 19, 41, 48, 56, 72, 104, 118], "demand": [11], "for": [12, 65, 173], "more": [13], "efficient": [14], "optimization": [15], "techniques.": [16], "Among": [17], "these,": [18], "Lookahead": [20, 116], "family": [21], "optimizers": [23], "employs": [24], "a": [25, 45, 60, 84, 110, 137, 142, 170, 174], "two-loop": [26], "framework,": [27], "maintaining": [28], "fast": [29, 42], "and": [30, 160, 163, 181], "slow": [31, 57], "sets": [32], "model": [34, 153, 166], "weights.": [35, 58], "Multiple": [36], "inner": [37, 177], "optimizer": [38], "steps": [39], "on": [40], "weights": [43], "produce": [44], "trajectory": [46], "-": [47, 50, 134], "pseudo-gradient": [49, 74], "that": [51, 93, 126, 150], "is": [52, 169], "used": [53], "to": [54, 71, 79, 103, 141], "update": [55], "DiLoCo,": [59], "notable": [61], "example": [62], "originally": [63], "designed": [64], "distributed": [66], "training,": [67], "applies": [68], "Nesterov": [69, 101, 120], "momentum": [70, 102], "averaged": [73], "from": [75, 99], "multiple": [76], "workers,": [77], "claiming": [78], "even": [80], "outperform": [81], "AdamW": [82, 180], "in": [83, 109, 136], "non-distributed": [85, 111, 138], "setup.": [86], "In": [87], "this": [88, 115], "paper,": [89], "we": [90], "empirically": [91], "show": [92], "DiLoCo's": [94], "surprising": [95], "effectiveness": [96], "stems": [97], "primarily": [98], "applying": [100], "pseudo-gradient,": [105], "which": [106], "improves": [107], "training": [108, 146], "setting.": [112], "We": [113, 124], "call": [114], "variant": [117], "Step-$K$": [119], "Outer": [121], "Optimizer": [122], "(SNOO).": [123], "demonstrate": [125], "SNOO": [127, 168], "achieves": [128], "compute": [129, 159], "factor": [130], "gains": [131], "1.5": [133], "2.5$\\times$": [135], "setting": [139], "up": [140], "scale": [143], "1e23": [145], "FLOPs,": [147], "with": [148, 152, 165], "improvements": [149], "increase": [151], "size.": [154], "Because": [155], "its": [157], "minimal": [158], "memory": [161], "overhead": [162], "compatibility": [164], "sharding,": [167], "practical": [171], "enhancement": [172], "variety": [175], "optimizers,": [178], "including": [179], "Muon.": [182]}, "primary_topic": {"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.9970999956130981, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T11612", "display_name": "Stochastic Gradient Optimization Techniques", "score": 0.9970999956130981, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T12535", "display_name": "Machine Learning and Data Classification", "score": 0.0003000000142492354, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10028", "display_name": "Topic Modeling", "score": 0.0003000000142492354, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2510.15830", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2510.15830", "pdf_url": "https://arxiv.org/pdf/2510.15830", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W7117692124", "doi": "https://doi.org/10.48550/arxiv.2512.22180", "title": "iOS as Acceleration", "publication_year": 2025, "publication_date": "2025-12-19", "type": "preprint", "language": null, "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5101557378", "display_name": "Alexander Chen", "orcid": "https://orcid.org/0000-0003-2801-9566", "observed_orcids": ["https://orcid.org/0000-0003-2801-9566"]}, "institutions": [], "countries": [], "is_corresponding": true, "raw_author_name": "Chen, Alexander K.", "raw_affiliation_strings": [], "raw_orcid": null, "affiliations": []}], "abstract_inverted_index": {"Practical": [0], "utilization": [1], "of": [2, 80, 159], "large-scale": [3], "machine": [4, 176], "learning": [5], "requires": [6], "a": [7, 11, 15, 33, 112, 116, 132], "powerful": [8, 95], "compute": [9, 71, 134], "setup,": [10], "necessity": [12], "which": [13], "poses": [14], "significant": [16, 129], "barrier": [17], "to": [18, 35, 67, 119, 171, 175], "engagement": [19], "with": [20, 93], "such": [21], "artificial": [22], "intelligence": [23], "in": [24, 131], "more": [25], "restricted": [26], "system": [27, 114], "environments.": [28], "While": [29], "cloud": [30], "computing": [31, 61], "offers": [32], "solution": [34], "weaker": [36, 69], "local": [37, 70], "environments,": [38], "certain": [39], "situations": [40], "like": [41, 102], "training": [42], "involving": [43], "private": [44], "or": [45, 55], "sensitive": [46], "data,": [47], "physical": [48], "environments": [49], "not": [50], "available": [51], "through": [52], "the": [53, 65, 163, 166], "cloud,": [54], "higher": [56], "anticipated": [57], "usage": [58], "costs,": [59], "necessitate": [60], "locally.": [62], "We": [63, 110, 147], "explore": [64], "potential": [66, 164], "improve": [68], "systems": [72], "at": [73], "zero": [74], "additional": [75], "cost": [76], "by": [77, 136], "taking": [78], "advantage": [79], "ubiquitous": [81], "yet": [82], "underutilized": [83], "resources:": [84], "mobile": [85, 169], "phones.": [86], "Specifically,": [87], "recent": [88], "iOS": [89, 122], "phones": [90], "are": [91], "equipped": [92], "surprisingly": [94], "processors,": [96], "but": [97], "they": [98], "also": [99], "face": [100], "limitations": [101], "memory": [103], "constraints,": [104], "thermal": [105], "throttling,": [106], "and": [107, 143, 152], "OS": [108], "sandboxing.": [109], "present": [111], "proof-of-concept": [113], "demonstrating": [115], "novel": [117], "approach": [118], "harness": [120], "an": [121], "device": [123], "via": [124], "distributed": [125], "pipeline": [126], "parallelism,": [127], "achieving": [128], "benefits": [130], "lesser": [133], "environment": [135], "accelerating": [137], "modest": [138], "model": [139], "training,": [140], "batch": [141], "inference,": [142], "agentic": [144], "LRM": [145], "tool-usage.": [146], "discuss": [148], "practical": [149], "use-cases,": [150], "limitations,": [151], "directions": [153], "for": [154, 165], "future": [155], "work.": [156], "The": [157], "findings": [158], "this": [160], "paper": [161], "highlight": [162], "improving": [167], "commonplace": [168], "devices": [170], "provide": [172], "greater": [173], "contributions": [174], "learning.": [177]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.6147000193595886, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.6147000193595886, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10273", "display_name": "IoT and Edge/Fog Computing", "score": 0.13989999890327454, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.05460000038146973, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "pmh:oai:arXiv.org:2512.22180", "is_oa": true, "landing_page_url": "http://arxiv.org/abs/2512.22180", "pdf_url": "https://arxiv.org/pdf/2512.22180", "source": {"id": "https://openalex.org/S4306400194", "display_name": "arXiv (Cornell University)", "issn_l": "2331-8422", "issn": ["2331-8422"], "is_oa": true, "is_in_doaj": false, "is_core": false, "listed_in": [], "host_organization": "https://openalex.org/I205783295", "host_organization_name": "Cornell University", "host_organization_lineage": ["https://openalex.org/I205783295"], "host_organization_lineage_names": [], "type": "repository"}, "license": null, "license_id": null, "version": "submittedVersion", "is_accepted": false, "is_published": false, "raw_source_name": null, "raw_type": "text"}, "cited_by_count": 0, "is_retracted": false}, {"id": "https://openalex.org/W4415970885", "doi": "https://doi.org/10.1109/tpds.2025.3629667", "title": "S-Leon: An Efficient Split Learning Framework Over Heterogeneous LEO Satellite Networks", "publication_year": 2025, "publication_date": "2025-11-06", "type": "article", "language": null, "authorships": [{"author_position": "first", "author": {"id": "https://openalex.org/A5056773892", "display_name": "Y. Zhang", "orcid": "https://orcid.org/0000-0001-7349-3780", "observed_orcids": ["https://orcid.org/0000-0001-7349-3780"]}, "institutions": [{"id": "https://openalex.org/I24943067", "display_name": "Fudan University", "ror": "https://ror.org/013q1eq08", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I24943067"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yuxin Zhang", "raw_affiliation_strings": ["Institute of Space Internet, Fudan University, Shanghai, China"], "raw_orcid": "https://orcid.org/0000-0001-7349-3780", "affiliations": [{"raw_affiliation_string": "Institute of Space Internet, Fudan University, Shanghai, China", "institution_ids": ["https://openalex.org/I24943067"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5100656819", "display_name": "Zhe Chen", "orcid": "https://orcid.org/0000-0002-3215-2696", "observed_orcids": ["https://orcid.org/0000-0002-3215-2696", "https://orcid.org/0000-0003-2286-4988"]}, "institutions": [{"id": "https://openalex.org/I24943067", "display_name": "Fudan University", "ror": "https://ror.org/013q1eq08", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I24943067"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Zhe Chen", "raw_affiliation_strings": ["Institute of Space Internet, Fudan University, Shanghai, China"], "raw_orcid": "https://orcid.org/0000-0002-3215-2696", "affiliations": [{"raw_affiliation_string": "Institute of Space Internet, Fudan University, Shanghai, China", "institution_ids": ["https://openalex.org/I24943067"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5022699488", "display_name": "Xuanjie Hu", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I24943067", "display_name": "Fudan University", "ror": "https://ror.org/013q1eq08", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I24943067"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Xuanjie Hu", "raw_affiliation_strings": ["Institute of Space Internet, Fudan University, Shanghai, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "Institute of Space Internet, Fudan University, Shanghai, China", "institution_ids": ["https://openalex.org/I24943067"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5003998216", "display_name": "Jin Zhao", "orcid": "https://orcid.org/0000-0002-9807-2648", "observed_orcids": ["https://orcid.org/0000-0002-9807-2648"]}, "institutions": [{"id": "https://openalex.org/I24943067", "display_name": "Fudan University", "ror": "https://ror.org/013q1eq08", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I24943067"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Jin Zhao", "raw_affiliation_strings": ["Institute of Space Internet, Fudan University, Shanghai, China"], "raw_orcid": "https://orcid.org/0000-0002-9807-2648", "affiliations": [{"raw_affiliation_string": "Institute of Space Internet, Fudan University, Shanghai, China", "institution_ids": ["https://openalex.org/I24943067"]}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5075109962", "display_name": "Yue Gao", "orcid": "https://orcid.org/0000-0001-6502-9910", "observed_orcids": ["https://orcid.org/0000-0001-6502-9910", "https://orcid.org/0000-0002-6027-5618"]}, "institutions": [{"id": "https://openalex.org/I24943067", "display_name": "Fudan University", "ror": "https://ror.org/013q1eq08", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I24943067"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yue Gao", "raw_affiliation_strings": ["Institute of Space Internet, Fudan University, Shanghai, China"], "raw_orcid": "https://orcid.org/0000-0001-6502-9910", "affiliations": [{"raw_affiliation_string": "Institute of Space Internet, Fudan University, Shanghai, China", "institution_ids": ["https://openalex.org/I24943067"]}]}], "abstract_inverted_index": {"The": [0], "rapid": [1], "deployment": [2], "of": [3, 96, 165], "low": [4], "Earth": [5], "orbit": [6], "(LEO)": [7], "satellite": [8, 97, 119, 124, 145, 152, 184, 191], "systems": [9], "has": [10, 69], "propelled": [11], "various": [12], "space-based": [13], "applications": [14], "(e.g.,": [15], "agricultural": [16], "monitoring": [17], "and": [18, 49, 94, 134, 162], "disaster": [19], "response),": [20], "which": [21], "increasingly": [22], "rely": [23], "on": [24, 62, 89], "advancements": [25], "in": [26], "deep": [27], "learning": [28, 67], "(DL).": [29], "However,": [30], "ground": [31, 141], "stations": [32], "(GS)": [33], "cannot": [34], "download": [35], "such": [36], "massive": [37], "raw": [38, 87], "data": [39, 88], "for": [40], "centralized": [41], "training": [42, 61, 78, 129, 174, 180], "due": [43], "to": [44, 59, 75, 80, 112, 127, 139, 176], "intermittent": [45], "connectivity": [46, 93], "between": [47], "satellites": [48], "GS,": [50], "while": [51, 85], "the": [52, 159, 178], "scaled-up": [53], "DL": [54], "models": [55], "pose": [56], "substantial": [57, 100], "barriers": [58], "distributed": [60], "resource-constrained": [63], "satellites.": [64, 167], "Although": [65], "split": [66], "(SL)": [68], "emerged": [70], "as": [71], "a": [72, 123, 151, 171], "promising": [73], "solution": [74], "offload": [76], "major": [77], "workloads": [79], "GS": [81], "via": [82], "model": [83, 126, 153], "partitioning": [84], "retaining": [86], "satellites,": [90], "limited": [91], "satellite-GS": [92], "heterogeneity": [95], "resources": [98], "remain": [99], "barriers.": [101], "In": [102], "this": [103], "paper,": [104], "we": [105, 149, 169], "propose": [106], "S-Leon,": [107], "an": [108], "SL": [109], "framework": [110], "tailored": [111], "tackle": [113], "these": [114], "challenges": [115], "within": [116], "heterogeneous": [117, 160], "LEO": [118, 190], "networks.": [120], "We": [121], "develop": [122, 170], "early-exit": [125], "eliminate": [128], "disruptions": [130], "during": [131], "non-contact": [132], "periods": [133], "employ": [135], "online": [136], "knowledge": [137], "distillation": [138], "incorporate": [140], "knowledge,": [142], "further": [143], "enhancing": [144], "local": [146], "training.": [147], "Moreover,": [148], "devise": [150], "customization": [154], "method": [155], "that": [156, 194], "simultaneously": [157], "accommodates": [158], "computation": [161], "communication": [163], "capabilities": [164], "individual": [166], "Lastly,": [168], "partial": [172], "model-agnostic": [173], "strategy": [175], "optimize": [177], "collaborative": [179], "effectiveness": [181], "across": [182], "customized": [183], "models.": [185], "Extensive": [186], "experiments": [187], "with": [188], "real-world": [189], "networks": [192], "demonstrate": [193], "S-Leon": [195], "outperforms": [196], "state-of-the-art": [197], "benchmarks.": [198]}, "primary_topic": {"id": "https://openalex.org/T12042", "display_name": "Satellite Communication Systems", "score": 0.9496999979019165, "subfield": {"id": "https://openalex.org/subfields/2202", "display_name": "Aerospace Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T12042", "display_name": "Satellite Communication Systems", "score": 0.9496999979019165, "subfield": {"id": "https://openalex.org/subfields/2202", "display_name": "Aerospace Engineering"}, "field": {"id": "https://openalex.org/fields/22", "display_name": "Engineering"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10764", "display_name": "Privacy-Preserving Technologies in Data", "score": 0.02449999935925007, "subfield": {"id": "https://openalex.org/subfields/1702", "display_name": "Artificial Intelligence"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10036", "display_name": "Advanced Neural Network Applications", "score": 0.002199999988079071, "subfield": {"id": "https://openalex.org/subfields/1707", "display_name": "Computer Vision and Pattern Recognition"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/tpds.2025.3629667", "is_oa": false, "landing_page_url": "https://doi.org/10.1109/tpds.2025.3629667", "pdf_url": null, "source": {"id": "https://openalex.org/S97130795", "display_name": "IEEE Transactions on Parallel and Distributed Systems", "issn_l": "1045-9219", "issn": ["1045-9219", "1558-2183", "2161-9883"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-2", "norway-2"], "host_organization": "https://openalex.org/P4310319808", "host_organization_name": "Institute of Electrical and Electronics Engineers", "host_organization_lineage": ["https://openalex.org/P4310319808"], "host_organization_lineage_names": ["Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Transactions on Parallel and Distributed Systems", "raw_type": "journal-article"}, "cited_by_count": 1, "is_retracted": false}, {"id": "https://openalex.org/W4411639943", "doi": "https://doi.org/10.1109/tpds.2025.3583165", "title": "Oases: Efficient Large-Scale Model Training on Commodity Servers via Overlapped and Automated Tensor Model Parallelism", "publication_year": 2025, "publication_date": "2025-06-25", "type": "article", "language": "en", "authorships": [{"author_position": "first", "author": {"id": null, "display_name": "Shengwei Li", "orcid": "https://orcid.org/0000-0002-7419-1511", "observed_orcids": ["https://orcid.org/0000-0002-7419-1511"]}, "institutions": [{"id": "https://openalex.org/I170215575", "display_name": "National University of Defense Technology", "ror": "https://ror.org/05d2yfz11", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I170215575"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Shengwei Li", "raw_affiliation_strings": ["National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China", "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology in Changsha, Hunan, China"], "raw_orcid": "https://orcid.org/0000-0002-7419-1511", "affiliations": [{"raw_affiliation_string": "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China", "institution_ids": ["https://openalex.org/I170215575"]}, {"raw_affiliation_string": "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology in Changsha, Hunan, China", "institution_ids": ["https://openalex.org/I170215575"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5055426388", "display_name": "Zhiquan Lai", "orcid": "https://orcid.org/0000-0002-3458-4732", "observed_orcids": ["https://orcid.org/0000-0002-3458-4732"]}, "institutions": [{"id": "https://openalex.org/I170215575", "display_name": "National University of Defense Technology", "ror": "https://ror.org/05d2yfz11", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I170215575"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Zhiquan Lai", "raw_affiliation_strings": ["National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China", "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology in Changsha, Hunan, China"], "raw_orcid": "https://orcid.org/0000-0002-3458-4732", "affiliations": [{"raw_affiliation_string": "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China", "institution_ids": ["https://openalex.org/I170215575"]}, {"raw_affiliation_string": "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology in Changsha, Hunan, China", "institution_ids": ["https://openalex.org/I170215575"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Dongsheng Li", "orcid": "https://orcid.org/0000-0001-9743-2034", "observed_orcids": ["https://orcid.org/0000-0001-9743-2034"]}, "institutions": [{"id": "https://openalex.org/I170215575", "display_name": "National University of Defense Technology", "ror": "https://ror.org/05d2yfz11", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I170215575"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Dongsheng Li", "raw_affiliation_strings": ["National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China", "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology in Changsha, Hunan, China"], "raw_orcid": "https://orcid.org/0000-0001-9743-2034", "affiliations": [{"raw_affiliation_string": "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China", "institution_ids": ["https://openalex.org/I170215575"]}, {"raw_affiliation_string": "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology in Changsha, Hunan, China", "institution_ids": ["https://openalex.org/I170215575"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5101206777", "display_name": "Yanqi Hao", "orcid": null, "observed_orcids": []}, "institutions": [{"id": "https://openalex.org/I170215575", "display_name": "National University of Defense Technology", "ror": "https://ror.org/05d2yfz11", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I170215575"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Yanqi Hao", "raw_affiliation_strings": ["National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China", "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology in Changsha, Hunan, China"], "raw_orcid": null, "affiliations": [{"raw_affiliation_string": "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China", "institution_ids": ["https://openalex.org/I170215575"]}, {"raw_affiliation_string": "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology in Changsha, Hunan, China", "institution_ids": ["https://openalex.org/I170215575"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Weijie Liu", "orcid": "https://orcid.org/0000-0002-0354-9467", "observed_orcids": ["https://orcid.org/0000-0002-0354-9467"]}, "institutions": [{"id": "https://openalex.org/I170215575", "display_name": "National University of Defense Technology", "ror": "https://ror.org/05d2yfz11", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I170215575"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Weijie Liu", "raw_affiliation_strings": ["National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China", "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology in Changsha, Hunan, China"], "raw_orcid": "https://orcid.org/0000-0002-0354-9467", "affiliations": [{"raw_affiliation_string": "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China", "institution_ids": ["https://openalex.org/I170215575"]}, {"raw_affiliation_string": "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology in Changsha, Hunan, China", "institution_ids": ["https://openalex.org/I170215575"]}]}, {"author_position": "middle", "author": {"id": "https://openalex.org/A5074708780", "display_name": "Keshi Ge", "orcid": "https://orcid.org/0000-0002-0669-6892", "observed_orcids": ["https://orcid.org/0000-0002-0669-6892"]}, "institutions": [{"id": "https://openalex.org/I170215575", "display_name": "National University of Defense Technology", "ror": "https://ror.org/05d2yfz11", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I170215575"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Keshi Ge", "raw_affiliation_strings": ["National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China", "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology in Changsha, Hunan, China"], "raw_orcid": "https://orcid.org/0000-0002-0669-6892", "affiliations": [{"raw_affiliation_string": "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China", "institution_ids": ["https://openalex.org/I170215575"]}, {"raw_affiliation_string": "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology in Changsha, Hunan, China", "institution_ids": ["https://openalex.org/I170215575"]}]}, {"author_position": "middle", "author": {"id": null, "display_name": "Xiaoge Deng", "orcid": "https://orcid.org/0000-0003-0622-1202", "observed_orcids": ["https://orcid.org/0000-0003-0622-1202"]}, "institutions": [], "countries": [], "is_corresponding": false, "raw_author_name": "Xiaoge Deng", "raw_affiliation_strings": ["Intelligent Game and Decision Lab, Beijing, China"], "raw_orcid": "https://orcid.org/0000-0003-0622-1202", "affiliations": [{"raw_affiliation_string": "Intelligent Game and Decision Lab, Beijing, China", "institution_ids": []}]}, {"author_position": "last", "author": {"id": "https://openalex.org/A5100625354", "display_name": "Kai L\u00fc", "orcid": "https://orcid.org/0000-0002-6378-7002", "observed_orcids": ["https://orcid.org/0000-0002-6378-7002", "https://orcid.org/0000-0003-2284-7897", "https://orcid.org/0000-0002-7757-4083", "https://orcid.org/0000-0003-4907-306X"]}, "institutions": [{"id": "https://openalex.org/I170215575", "display_name": "National University of Defense Technology", "ror": "https://ror.org/05d2yfz11", "country_code": "CN", "type": "education", "lineage": ["https://openalex.org/I170215575"]}], "countries": ["CN"], "is_corresponding": false, "raw_author_name": "Kai Lu", "raw_affiliation_strings": ["National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China", "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology in Changsha, Hunan, China"], "raw_orcid": "https://orcid.org/0000-0002-6378-7002", "affiliations": [{"raw_affiliation_string": "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China", "institution_ids": ["https://openalex.org/I170215575"]}, {"raw_affiliation_string": "National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology in Changsha, Hunan, China", "institution_ids": ["https://openalex.org/I170215575"]}]}], "abstract_inverted_index": {"Deep": [0], "learning": [1], "is": [2, 12, 56, 140], "experiencing": [3], "a": [4, 99], "rise": [5], "in": [6], "large-scale": [7, 10, 18, 91], "models.": [8], "Training": [9], "models": [11, 19], "costly,": [13], "prompting": [14], "researchers": [15, 25], "to": [16, 89, 104, 131, 142, 164, 186], "train": [17], "on": [20, 44, 94, 153], "commodity": [21, 95, 159], "servers": [22], "that": [23, 110, 120, 171], "more": [24], "can": [26], "access.": [27], "The": [28], "massive": [29], "number": [30], "of": [31, 36, 129, 146, 175, 184], "parameters": [32], "necessitates": [33], "the": [34, 51, 59, 73, 117, 123, 144, 179], "use": [35], "model": [37, 48, 53, 60, 92, 125, 143, 155], "parallelism": [38, 54], "training": [39, 45, 74, 93, 101], "methods.": [40], "Existing": [41], "studies": [42], "focus": [43], "with": [46, 86], "pipeline": [47], "parallelism.": [49], "However,": [50], "tensor": [52], "(TMP)": [55], "inevitable": [57], "when": [58], "size": [61], "keeps": [62], "increasing,": [63], "where": [64], "frequent": [65], "data-dependent": [66], "communication": [67, 88, 107], "and": [68, 108, 157, 161, 182], "computation": [69, 109], "operations": [70], "significantly": [71], "reduce": [72], "efficiency.": [75], "In": [76], "this": [77], "paper,": [78], "we": [79, 115], "present": [80], "Oases,": [81], "an": [82], "automated": [83], "TMP": [84, 130], "method": [85], "overlapped": [87, 147], "accelerate": [90], "servers.": [96], "Oases": [97, 118, 138, 152, 163, 172], "proposes": [98], "fine-grained": [100], "operation": [102], "schedule": [103], "maximize": [105], "overlapping": [106], "have": [111], "data": [112], "dependence.": [113], "Additionally,": [114], "design": [116], "planner": [119, 139], "searches": [121], "for": [122], "best": [124], "parameter": [126], "partition": [127], "strategy": [128], "achieve": [132], "further": [133], "accelerations.": [134], "Unlike": [135], "existing": [136], "methods,": [137], "tailored": [141], "cost": [145], "communication-computation": [148], "operations.": [149], "We": [150], "evaluate": [151], "various": [154], "settings": [156], "two": [158], "clusters,": [160], "compare": [162], "four": [165], "state-of-the-art": [166], "implementations.": [167], "Experimental": [168], "results": [169], "show": [170], "achieves": [173], "speedups": [174, 183], "1.01\u20131.48": [176], "\u00d7": [177, 188], "over": [178, 189], "fastest": [180], "baseline,": [181], "up": [185], "1.95": [187], "Megatron.": [190]}, "primary_topic": {"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9570000171661377, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, "topics": [{"id": "https://openalex.org/T10054", "display_name": "Parallel Computing and Optimization Techniques", "score": 0.9570000171661377, "subfield": {"id": "https://openalex.org/subfields/1708", "display_name": "Hardware and Architecture"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10715", "display_name": "Distributed and Parallel Computing Systems", "score": 0.004800000227987766, "subfield": {"id": "https://openalex.org/subfields/1705", "display_name": "Computer Networks and Communications"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}, {"id": "https://openalex.org/T10101", "display_name": "Cloud Computing and Resource Management", "score": 0.004800000227987766, "subfield": {"id": "https://openalex.org/subfields/1710", "display_name": "Information Systems"}, "field": {"id": "https://openalex.org/fields/17", "display_name": "Computer Science"}, "domain": {"id": "https://openalex.org/domains/3", "display_name": "Physical Sciences"}}], "primary_location": {"id": "doi:10.1109/tpds.2025.3583165", "is_oa": false, "landing_page_url": "https://doi.org/10.1109/tpds.2025.3583165", "pdf_url": null, "source": {"id": "https://openalex.org/S97130795", "display_name": "IEEE Transactions on Parallel and Distributed Systems", "issn_l": "1045-9219", "issn": ["1045-9219", "1558-2183", "2161-9883"], "is_oa": false, "is_in_doaj": false, "is_core": true, "listed_in": ["cwts-core", "jufo-2", "norway-2"], "host_organization": "https://openalex.org/P4310319808", "host_organization_name": "Institute of Electrical and Electronics Engineers", "host_organization_lineage": ["https://openalex.org/P4310319808"], "host_organization_lineage_names": ["Institute of Electrical and Electronics Engineers"], "type": "journal"}, "license": null, "license_id": null, "version": "publishedVersion", "is_accepted": true, "is_published": true, "raw_source_name": "IEEE Transactions on Parallel and Distributed Systems", "raw_type": "journal-article"}, "cited_by_count": 2, "is_retracted": false}], "group_by": []}