{"id":"https://openalex.org/W4416251999","doi":"https://doi.org/10.1109/ijcnn64981.2025.11228828","title":"DYNAMAX: Dynamic computing for Transformers and Mamba based architectures","display_name":"DYNAMAX: Dynamic computing for Transformers and Mamba based architectures","publication_year":2025,"publication_date":"2025-06-30","ids":{"openalex":"https://openalex.org/W4416251999","doi":"https://doi.org/10.1109/ijcnn64981.2025.11228828"},"language":"en","primary_location":{"id":"doi:10.1109/ijcnn64981.2025.11228828","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn64981.2025.11228828","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 International Joint Conference on Neural Networks (IJCNN)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5120354550","display_name":"Miguel Nogales","orcid":null},"institutions":[{"id":"https://openalex.org/I57201433","display_name":"Universit\u00e0 della Svizzera italiana","ror":"https://ror.org/03c4atk17","country_code":"CH","type":"education","lineage":["https://openalex.org/I57201433"]}],"countries":["CH"],"is_corresponding":true,"raw_author_name":"Miguel Nogales","raw_affiliation_strings":["Università Della Svizzera Italiana,Lugano,Switzerland"],"affiliations":[{"raw_affiliation_string":"Università Della Svizzera Italiana,Lugano,Switzerland","institution_ids":["https://openalex.org/I57201433"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008809497","display_name":"Matteo Gambella","orcid":"https://orcid.org/0000-0001-9734-670X"},"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":"Matteo Gambella","raw_affiliation_strings":["Politecnico di Milano,Milano,Italy"],"affiliations":[{"raw_affiliation_string":"Politecnico di Milano,Milano,Italy","institution_ids":["https://openalex.org/I93860229"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5035547226","display_name":"Manuel Roveri","orcid":"https://orcid.org/0000-0001-7828-7687"},"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":"Manuel Roveri","raw_affiliation_strings":["Politecnico di Milano,Milano,Italy"],"affiliations":[{"raw_affiliation_string":"Politecnico di Milano,Milano,Italy","institution_ids":["https://openalex.org/I93860229"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5120354550"],"corresponding_institution_ids":["https://openalex.org/I57201433"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.36688345,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"8"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10904","display_name":"Embedded Systems Design Techniques","score":0.12620000541210175,"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.12620000541210175,"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/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.056299999356269836,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.05510000139474869,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.6251999735832214},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6001999974250793},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.49300000071525574},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.45969998836517334},{"id":"https://openalex.org/keywords/adaptability","display_name":"Adaptability","score":0.4115999937057495}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6776000261306763},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.6251999735832214},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6001999974250793},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.49300000071525574},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.45969998836517334},{"id":"https://openalex.org/C177606310","wikidata":"https://www.wikidata.org/wiki/Q5674297","display_name":"Adaptability","level":2,"score":0.4115999937057495},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3578000068664551},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.31310001015663147},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.31139999628067017},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.30970001220703125},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3034999966621399},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.30320000648498535},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.2669999897480011},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2667999863624573},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.257999986410141}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/ijcnn64981.2025.11228828","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn64981.2025.11228828","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 International Joint Conference on Neural Networks (IJCNN)","raw_type":"proceedings-article"},{"id":"pmh:oai:re.public.polimi.it:11311/1295445","is_oa":false,"landing_page_url":"https://hdl.handle.net/11311/1295445","pdf_url":null,"source":{"id":"https://openalex.org/S4306400312","display_name":"Virtual Community of Pathological Anatomy (University of Castilla La Mancha)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I79189158","host_organization_name":"University of Castilla-La Mancha","host_organization_lineage":["https://openalex.org/I79189158"],"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":"info:eu-repo/semantics/conferenceObject"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":18,"referenced_works":["https://openalex.org/W2108278040","https://openalex.org/W2110798204","https://openalex.org/W2962677625","https://openalex.org/W2963339397","https://openalex.org/W2963393494","https://openalex.org/W2964223283","https://openalex.org/W2979826702","https://openalex.org/W3035030897","https://openalex.org/W3196295870","https://openalex.org/W3201174429","https://openalex.org/W4312849330","https://openalex.org/W4385245566","https://openalex.org/W4385482751","https://openalex.org/W4385958936","https://openalex.org/W4389519070","https://openalex.org/W4389523983","https://openalex.org/W4401726555","https://openalex.org/W4414281281"],"related_works":[],"abstract_inverted_index":{"Early":[0],"exits":[1],"(EEs)":[2],"offer":[3],"a":[4,18,23,46,141],"promising":[5],"approach":[6],"to":[7,38,66,111,126],"reducing":[8],"computational":[9,128,150],"costs":[10],"and":[11,97,124,131,145,152,171,177],"latency":[12],"by":[13],"dynamically":[14],"terminating":[15],"inference":[16,173],"once":[17],"satisfactory":[19],"prediction":[20],"confidence":[21],"on":[22],"data":[24,118],"sample":[25],"is":[26],"achieved.":[27],"Although":[28],"many":[29],"works":[30],"integrate":[31,81],"EEs":[32,82],"into":[33,83],"encoder-only":[34],"Transformers,":[35],"their":[36],"application":[37],"decoder-only":[39],"architectures":[40,51,73],"and,":[41],"more":[42],"importantly,":[43],"Mamba":[44,72,84,88,115,139,187],"models,":[45],"novel":[47],"family":[48],"of":[49,71,138,186],"state-space":[50],"in":[52,148,174,188],"the":[53,63,68,106,112,136,183],"LLM":[54],"realm,":[55],"remains":[56],"insufficiently":[57],"explored.":[58],"This":[59,180],"work":[60],"introduces":[61],"DYNAMAX,":[62],"first":[64],"framework":[65],"exploit":[67],"unique":[69],"properties":[70],"for":[74,94,163,169,193],"early":[75],"exit":[76],"mechanisms.":[77],"We":[78],"not":[79],"only":[80],"but":[85],"also":[86],"repurpose":[87],"as":[89,121,140],"an":[90],"efficient":[91,172],"EE":[92,143],"classifier":[93,144],"both":[95],"Mamba-based":[96],"transformer-based":[98],"LLMs,":[99],"showcasing":[100],"its":[101,146],"versatility.":[102],"Our":[103],"experiments":[104],"employ":[105],"Mistral":[107],"7B":[108,114],"transformer":[109],"compared":[110],"Codestral":[113],"model,":[116],"using":[117],"sets":[119],"such":[120],"TruthfulQA,":[122],"CoQA,":[123],"TriviaQA":[125],"evaluate":[127],"savings,":[129],"accuracy,":[130],"consistency.":[132],"The":[133],"results":[134],"highlight":[135],"adaptability":[137],"powerful":[142],"efficiency":[147],"balancing":[149],"cost":[151],"performance":[153],"quality":[154],"across":[155],"NLP":[156],"tasks.":[157],"By":[158],"leveraging":[159],"Mamba\u2019s":[160],"inherent":[161],"design":[162],"dynamic":[164,190],"processing,":[165],"we":[166],"open":[167],"pathways":[168],"scalable":[170],"embedded":[175],"applications":[176],"resource-constrained":[178],"environments.":[179],"study":[181],"underscores":[182],"transformative":[184],"potential":[185],"redefining":[189],"computing":[191],"paradigms":[192],"LLMs.":[194]},"counts_by_year":[],"updated_date":"2026-04-13T07:58:08.660418","created_date":"2025-10-10T00:00:00"}