{"id":"https://openalex.org/W7139122219","doi":"https://doi.org/10.1609/aaai.v40i39.40619","title":"MCW-KD: Multi-Cost Wasserstein Knowledge Distillation for Large Language Models","display_name":"MCW-KD: Multi-Cost Wasserstein Knowledge Distillation for Large Language Models","publication_year":2026,"publication_date":"2026-03-14","ids":{"openalex":"https://openalex.org/W7139122219","doi":"https://doi.org/10.1609/aaai.v40i39.40619"},"language":"en","primary_location":{"id":"doi:10.1609/aaai.v40i39.40619","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i39.40619","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.1609/aaai.v40i39.40619","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5121312474","display_name":"Hoang Tran Vuong","orcid":null},"institutions":[{"id":"https://openalex.org/I94518387","display_name":"Hanoi University of Science and Technology","ror":"https://ror.org/04nyv3z04","country_code":"VN","type":"education","lineage":["https://openalex.org/I94518387"]}],"countries":["VN"],"is_corresponding":false,"raw_author_name":"Hoang Tran Vuong","raw_affiliation_strings":["Hanoi University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hanoi University of Science and Technology","institution_ids":["https://openalex.org/I94518387"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014374360","display_name":"T. Kim Le","orcid":"https://orcid.org/0000-0001-5755-7473"},"institutions":[{"id":"https://openalex.org/I2801239119","display_name":"Australian Regenerative Medicine Institute","ror":"https://ror.org/02qa5kg76","country_code":"AU","type":"facility","lineage":["https://openalex.org/I2801037857","https://openalex.org/I2801239119","https://openalex.org/I56590836"]},{"id":"https://openalex.org/I56590836","display_name":"Monash University","ror":"https://ror.org/02bfwt286","country_code":"AU","type":"education","lineage":["https://openalex.org/I56590836"]},{"id":"https://openalex.org/I94518387","display_name":"Hanoi University of Science and Technology","ror":"https://ror.org/04nyv3z04","country_code":"VN","type":"education","lineage":["https://openalex.org/I94518387"]}],"countries":["AU","VN"],"is_corresponding":false,"raw_author_name":"Tue Le","raw_affiliation_strings":["Hanoi University of Science and Technology","Monash University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hanoi University of Science and Technology","institution_ids":["https://openalex.org/I94518387"]},{"raw_affiliation_string":"Monash University","institution_ids":["https://openalex.org/I2801239119","https://openalex.org/I56590836"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129887183","display_name":"Quyen Tran","orcid":null},"institutions":[{"id":"https://openalex.org/I102322142","display_name":"Rutgers, The State University of New Jersey","ror":"https://ror.org/05vt9qd57","country_code":"US","type":"education","lineage":["https://openalex.org/I102322142"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Quyen Tran","raw_affiliation_strings":["Rutgers University, New Jersey, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Rutgers University, New Jersey, USA","institution_ids":["https://openalex.org/I102322142"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125251729","display_name":"Linh Ngo Van","orcid":null},"institutions":[{"id":"https://openalex.org/I94518387","display_name":"Hanoi University of Science and Technology","ror":"https://ror.org/04nyv3z04","country_code":"VN","type":"education","lineage":["https://openalex.org/I94518387"]}],"countries":["VN"],"is_corresponding":false,"raw_author_name":"Linh Ngo Van","raw_affiliation_strings":["Hanoi University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hanoi University of Science and Technology","institution_ids":["https://openalex.org/I94518387"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5014374360","display_name":"T. Kim Le","orcid":"https://orcid.org/0000-0001-5755-7473"},"institutions":[{"id":"https://openalex.org/I2801239119","display_name":"Australian Regenerative Medicine Institute","ror":"https://ror.org/02qa5kg76","country_code":"AU","type":"facility","lineage":["https://openalex.org/I2801037857","https://openalex.org/I2801239119","https://openalex.org/I56590836"]},{"id":"https://openalex.org/I56590836","display_name":"Monash University","ror":"https://ror.org/02bfwt286","country_code":"AU","type":"education","lineage":["https://openalex.org/I56590836"]},{"id":"https://openalex.org/I94518387","display_name":"Hanoi University of Science and Technology","ror":"https://ror.org/04nyv3z04","country_code":"VN","type":"education","lineage":["https://openalex.org/I94518387"]}],"countries":["AU","VN"],"is_corresponding":false,"raw_author_name":"Trung Le","raw_affiliation_strings":["Hanoi University of Science and Technology","Monash University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hanoi University of Science and Technology","institution_ids":["https://openalex.org/I94518387"]},{"raw_affiliation_string":"Monash University","institution_ids":["https://openalex.org/I2801239119","https://openalex.org/I56590836"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.22921694,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"40","issue":"39","first_page":"33332","last_page":"33340"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.1273999959230423,"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":0.1273999959230423,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.08470000326633453,"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/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.07349999994039536,"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/distillation","display_name":"Distillation","score":0.6640999913215637},{"id":"https://openalex.org/keywords/lexical-analysis","display_name":"Lexical analysis","score":0.3887999951839447},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.38769999146461487},{"id":"https://openalex.org/keywords/mathematical-model","display_name":"Mathematical model","score":0.28220000863075256}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7332000136375427},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.6640999913215637},{"id":"https://openalex.org/C176982825","wikidata":"https://www.wikidata.org/wiki/Q835922","display_name":"Lexical analysis","level":2,"score":0.3887999951839447},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.38769999146461487},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3714999854564667},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.358599990606308},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3093000054359436},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3003999888896942},{"id":"https://openalex.org/C76969082","wikidata":"https://www.wikidata.org/wiki/Q486902","display_name":"Mathematical model","level":2,"score":0.28220000863075256},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.25619998574256897},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.2547999918460846}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1609/aaai.v40i39.40619","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i39.40619","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},{"id":"pmh:oai:ojs.aaai.org:article/40619","is_oa":false,"landing_page_url":"https://ojs.aaai.org/index.php/AAAI/article/view/40619","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"2159-5399","raw_type":"info:eu-repo/semantics/article"}],"best_oa_location":{"id":"doi:10.1609/aaai.v40i39.40619","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i39.40619","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education","score":0.7960572242736816}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Knowledge":[0,83],"distillation":[1],"(KD)":[2],"is":[3],"widely":[4],"recognized":[5],"as":[6],"an":[7],"effective":[8],"approach":[9],"for":[10,130],"compressing":[11],"large":[12],"language":[13],"models":[14,71,166],"(LLMs).":[15],"However,":[16],"standard":[17],"KD":[18,91,159],"methods":[19,55],"often":[20],"falter":[21],"when":[22,162],"confronted":[23],"with":[24,72],"architectural":[25],"or":[26],"tokenization":[27],"heterogeneity":[28],"between":[29,70],"teacher":[30,163],"and":[31,116,140,164],"student":[32,153,165],"models,":[33],"which":[34,62],"creates":[35],"a":[36,46,58,86,99,126],"mismatch":[37],"in":[38],"their":[39],"representations.":[40],"While":[41],"Optimal":[42],"Transport":[43],"(OT)":[44],"provides":[45],"promising":[47],"solution":[48],"to":[49,65,108,157],"align":[50,110],"these":[51],"representations,":[52],"most":[53],"OT-based":[54],"rely":[56],"on":[57,145],"single":[59],"cost":[60,96,106],"function,":[61],"isn\u2019t":[63],"enough":[64],"capture":[66],"the":[67,112,117,121,131],"multifaceted":[68],"discrepancies":[69],"distinct":[73],"designs.":[74],"To":[75],"address":[76],"this":[77],"limitation,":[78],"we":[79],"introduce":[80],"Multi-Cost":[81,133],"Wasserstein":[82,134],"Distillation":[84],"(MCW-KD),":[85],"novel":[87],"framework":[88],"that":[89,149],"enhances":[90],"by":[92],"simultaneously":[93],"optimizing":[94],"several":[95],"functions":[97],"within":[98],"unified":[100],"OT":[101],"formulation.":[102],"MCW-KD":[103,150],"employs":[104],"specific":[105],"matrices":[107],"effectively":[109],"both":[111,137],"final":[113],"hidden":[114],"states":[115],"output":[118],"distributions":[119],"of":[120],"models.":[122],"We":[123],"also":[124],"provide":[125],"rigorous":[127],"theoretical":[128],"foundation":[129],"proposed":[132],"Distance,":[135],"ensuring":[136],"mathematical":[138],"validity":[139],"computational":[141],"ability.":[142],"Extensive":[143],"experiments":[144],"instruction-following":[146],"datasets":[147],"demonstrate":[148],"significantly":[151],"improves":[152],"model":[154],"performance":[155],"compared":[156],"state-of-the-art":[158],"baselines,":[160],"especially":[161],"have":[167],"different":[168],"tokenizers.":[169]},"counts_by_year":[],"updated_date":"2026-08-26T07:47:46.906454","created_date":"2026-03-20T00:00:00"}