{"id":"https://openalex.org/W4416036115","doi":"https://doi.org/10.18653/v1/2025.emnlp-main.1097","title":"Continuous-Time Attention: PDE-Guided Mechanisms for Long-Sequence Transformers","display_name":"Continuous-Time Attention: PDE-Guided Mechanisms for Long-Sequence Transformers","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416036115","doi":"https://doi.org/10.18653/v1/2025.emnlp-main.1097"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.emnlp-main.1097","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.1097","pdf_url":"https://aclanthology.org/2025.emnlp-main.1097.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.emnlp-main.1097.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101539036","display_name":"Yukun Zhang","orcid":"https://orcid.org/0009-0004-0928-2136"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Yukun Zhang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5101538073","display_name":"Xueqing Zhou","orcid":"https://orcid.org/0000-0001-7654-6286"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xueqing Zhou","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5101539036"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.18407869,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"21654","last_page":"21674"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12611","display_name":"Neural Networks and Reservoir Computing","score":0.03680000081658363,"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/T12611","display_name":"Neural Networks and Reservoir Computing","score":0.03680000081658363,"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/T11948","display_name":"Machine Learning in Materials Science","score":0.03280000016093254,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.0284000001847744,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.38580000400543213},{"id":"https://openalex.org/keywords/troubleshooting","display_name":"Troubleshooting","score":0.24150000512599945},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.2337999939918518},{"id":"https://openalex.org/keywords/ferroresonance-in-electricity-networks","display_name":"Ferroresonance in electricity networks","score":0.23360000550746918}],"concepts":[{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.38580000400543213},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.31470000743865967},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.2944999933242798},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.260699987411499},{"id":"https://openalex.org/C147494362","wikidata":"https://www.wikidata.org/wiki/Q2078905","display_name":"Troubleshooting","level":2,"score":0.24150000512599945},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.2337999939918518},{"id":"https://openalex.org/C101382669","wikidata":"https://www.wikidata.org/wiki/Q1616769","display_name":"Ferroresonance in electricity networks","level":4,"score":0.23360000550746918},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.2223999947309494},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.20659999549388885},{"id":"https://openalex.org/C19474535","wikidata":"https://www.wikidata.org/wiki/Q131172","display_name":"Welding","level":2,"score":0.19910000264644623}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.emnlp-main.1097","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.1097","pdf_url":"https://aclanthology.org/2025.emnlp-main.1097.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.emnlp-main.1097","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.1097","pdf_url":"https://aclanthology.org/2025.emnlp-main.1097.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416036115.pdf","grobid_xml":"https://content.openalex.org/works/W4416036115.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0],"present":[1],"Continuous-Time":[2,80],"Attention,":[3],"a":[4,26,37,94,108],"novel":[5],"framework":[6],"that":[7,65,100],"infuses":[8],"partial":[9],"differential":[10],"equations":[11],"(PDEs)":[12],"into":[13,104],"the":[14,69,77],"Transformer's":[15],"attention":[16,28,32,67,105],"mechanism":[17],"to":[18,34],"better":[19],"handle":[20],"long":[21],"sequences.Instead":[22],"of":[23,72,96],"relying":[24],"on":[25],"static":[27],"matrix,":[29],"we":[30],"allow":[31],"weights":[33],"evolve":[35],"along":[36],"pseudo-time":[38],"dimension":[39],"governed":[40],"by":[41],"diffusion,":[42],"wave,":[43],"or":[44],"reaction-diffusion":[45],"dynamics.This":[46],"dynamic":[47],"process":[48],"systematically":[49],"smooths":[50],"local":[51],"noise,":[52],"strengthens":[53],"long-range":[54],"dependencies,":[55],"and":[56,75,89,115],"improves":[57,76],"gradient":[58],"stability":[59],"during":[60],"training.Our":[61],"theoretical":[62],"analysis":[63],"shows":[64],"PDE-driven":[66],"mitigates":[68],"exponential":[70],"decay":[71],"distant":[73],"interactions":[74],"optimization":[78],"landscape.Empirically,":[79],"Attention":[81],"achieves":[82],"consistent":[83],"performance":[84],"gains":[85],"over":[86],"both":[87],"standard":[88],"long-sequence":[90],"Transformer":[91,118],"variants":[92],"across":[93],"range":[95],"tasks.These":[97],"results":[98],"suggest":[99],"embedding":[101],"continuous-time":[102],"dynamics":[103],"mechanisms":[106],"is":[107],"promising":[109],"direction":[110],"for":[111],"enhancing":[112],"global":[113],"coherence":[114],"scalability":[116],"in":[117],"models.":[119]},"counts_by_year":[],"updated_date":"2026-03-11T06:11:40.159057","created_date":"2025-11-08T00:00:00"}