{"id":"https://openalex.org/W4412887820","doi":"https://doi.org/10.18653/v1/2025.findings-acl.1168","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","display_name":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412887820","doi":"https://doi.org/10.18653/v1/2025.findings-acl.1168"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.1168","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1168","pdf_url":"https://aclanthology.org/2025.findings-acl.1168.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":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.1168.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5008600285","display_name":"Miguel Romero Calvo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Miguel Romero Calvo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034588377","display_name":"Shuoyang Ding","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shuoyang Ding","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111092857","display_name":"Corey D Barrett","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Corey D Barrett","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090357305","display_name":"Georgiana Dinu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Georgiana Dinu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5082384108","display_name":"George Karypis","orcid":"https://orcid.org/0000-0003-2753-1437"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"George Karypis","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.14217264,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"22731","last_page":"22746"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.932200014591217,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.932200014591217,"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/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.9279000163078308,"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":0.9241999983787537,"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/task","display_name":"Task (project management)","score":0.7781350612640381},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7258332967758179},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.6790275573730469},{"id":"https://openalex.org/keywords/conditioning","display_name":"Conditioning","score":0.538110077381134},{"id":"https://openalex.org/keywords/task-analysis","display_name":"Task analysis","score":0.4386424124240875},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.357180118560791},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.11107218265533447},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.09839648008346558},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.07328629493713379},{"id":"https://openalex.org/keywords/systems-engineering","display_name":"Systems engineering","score":0.07126104831695557}],"concepts":[{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.7781350612640381},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7258332967758179},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.6790275573730469},{"id":"https://openalex.org/C45262634","wikidata":"https://www.wikidata.org/wiki/Q5159291","display_name":"Conditioning","level":2,"score":0.538110077381134},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.4386424124240875},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.357180118560791},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.11107218265533447},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.09839648008346558},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.07328629493713379},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.07126104831695557}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.1168","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1168","pdf_url":"https://aclanthology.org/2025.findings-acl.1168.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":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.1168","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1168","pdf_url":"https://aclanthology.org/2025.findings-acl.1168.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":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412887820.pdf","grobid_xml":"https://content.openalex.org/works/W4412887820.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W1564680838","https://openalex.org/W2003125260","https://openalex.org/W2060591604","https://openalex.org/W1992291644","https://openalex.org/W2166791242","https://openalex.org/W2585162246","https://openalex.org/W1934413089","https://openalex.org/W2098419343","https://openalex.org/W2051178964","https://openalex.org/W2989714914"],"abstract_inverted_index":{"Dense":[0],"embeddings":[1],"are":[2,106],"fundamental":[3,34],"to":[4,30,71,76],"modern":[5],"machine":[6],"learning":[7],"systems,":[8],"powering":[9],"Retrieval-Augmented":[10],"Generation":[11],"(RAG),":[12],"information":[13],"retrieval,":[14],"and":[15,51,94],"representation":[16],"learning.While":[17],"instruction-conditioning":[18],"has":[19],"become":[20],"the":[21,39,53,73],"dominant":[22],"approach":[23],"for":[24],"embedding":[25],"specialization,":[26],"its":[27],"direct":[28],"application":[29],"low-capacity":[31],"models":[32],"imposes":[33],"representational":[35],"constraints":[36],"that":[37,82],"limit":[38],"performance":[40,87,97],"gains":[41,88,98,105],"derived":[42],"from":[43],"specialization.In":[44],"this":[45],"paper,":[46],"we":[47],"analyze":[48],"these":[49,104],"limitations":[50],"introduce":[52],"Mixture":[54],"of":[55,117],"Task":[56],"Experts":[57],"(MoTE)":[58],"transformer":[59],"block,":[60],"which":[61],"leverages":[62],"task-specialized":[63],"parameters":[64],"trained":[65],"with":[66],"Task-Aware":[67],"Contrastive":[68],"Learning":[69],"(TA-CL)":[70],"enhance":[72],"model":[74],"ability":[75],"generate":[77],"specialized":[78],"embeddings.Empirical":[79],"results":[80],"show":[81],"MoTE":[83],"achieves":[84],"64%":[85],"higher":[86,96],"in":[89],"retrieval":[90],"datasets":[91,101],"(+3.27":[92],"+5.21)":[93],"43%":[95],"across":[99],"all":[100],"(+1.81":[102],"+2.60).Critically,":[103],"achieved":[107],"without":[108],"altering":[109],"instructions,":[110],"training":[111],"data,":[112],"inference":[113],"time,":[114],"or":[115],"number":[116],"active":[118],"parameters.":[119]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}