{"id":"https://openalex.org/W4411120616","doi":"https://doi.org/10.18653/v1/2025.findings-naacl.460","title":"MoDE: Effective Multi-task Parameter Efficient Fine-Tuning with a Mixture of Dyadic Experts","display_name":"MoDE: Effective Multi-task Parameter Efficient Fine-Tuning with a Mixture of Dyadic Experts","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4411120616","doi":"https://doi.org/10.18653/v1/2025.findings-naacl.460"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-naacl.460","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-naacl.460","pdf_url":"https://aclanthology.org/2025.findings-naacl.460.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: NAACL 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-naacl.460.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100635306","display_name":"Ning Lin","orcid":"https://orcid.org/0000-0002-6198-7043"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin Ning","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5038479755","display_name":"Harsh Lara","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Harsh Lara","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Meiqi Guo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Meiqi Guo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5087687501","display_name":"Abhinav Rastogi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Abhinav Rastogi","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":1.4143,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.78981797,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"8233","last_page":"8246"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.7939000129699707,"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/T10320","display_name":"Neural Networks and Applications","score":0.7939000129699707,"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/T10052","display_name":"Medical Image Segmentation Techniques","score":0.7748000025749207,"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/T10481","display_name":"Computer Graphics and Visualization Techniques","score":0.6965000033378601,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/mode","display_name":"Mode (computer interface)","score":0.7285942435264587},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6829370856285095},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6187050938606262},{"id":"https://openalex.org/keywords/human\u2013computer-interaction","display_name":"Human\u2013computer interaction","score":0.20736250281333923},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.11146625876426697},{"id":"https://openalex.org/keywords/systems-engineering","display_name":"Systems engineering","score":0.0800628662109375}],"concepts":[{"id":"https://openalex.org/C48677424","wikidata":"https://www.wikidata.org/wiki/Q6888088","display_name":"Mode (computer interface)","level":2,"score":0.7285942435264587},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6829370856285095},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6187050938606262},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.20736250281333923},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.11146625876426697},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.0800628662109375}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-naacl.460","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-naacl.460","pdf_url":"https://aclanthology.org/2025.findings-naacl.460.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: NAACL 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-naacl.460","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-naacl.460","pdf_url":"https://aclanthology.org/2025.findings-naacl.460.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: NAACL 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4411120616.pdf","grobid_xml":"https://content.openalex.org/works/W4411120616.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"Parameter-efficient":[0],"fine-tuning":[1,122],"techniques":[2],"like":[3],"Low-Rank":[4],"Adaptation":[5],"(LoRA)":[6],"have":[7,20],"revolutionized":[8],"the":[9,34,63,90,101],"adaptation":[10,141],"of":[11,23,37,46,107,111,135],"large":[12],"language":[13],"models":[14],"(LLMs)":[15],"to":[16,93,131],"diverse":[17,109],"tasks.Recent":[18],"efforts":[19],"explored":[21],"mixtures":[22],"LoRA":[24],"modules":[25],"for":[26,55,84,147],"multi-task":[27,57,120,139],"settings.However,":[28],"our":[29,42],"analysis":[30],"reveals":[31],"redundancy":[32],"in":[33,138],"down-projection":[35,64],"matrices":[36],"these":[38],"architectures.This":[39],"observation":[40],"motivates":[41],"proposed":[43],"method,":[44],"Mixture":[45],"Dyadic":[47],"Experts":[48],"(MoDE),":[49],"which":[50],"introduces":[51],"a":[52,108,132,144],"novel":[53],"design":[54,82],"efficient":[56],"adaptation.This":[58],"is":[59],"done":[60],"by":[61],"sharing":[62],"matrix":[65],"across":[66],"tasks":[67,113],"and":[68,114,142],"employing":[69],"atomic":[70],"rank-one":[71],"adapters,":[72],"coupled":[73],"with":[74],"routers":[75],"that":[76,116],"allow":[77],"more":[78,85],"sophisticated":[79],"task-level":[80],"specialization.Our":[81],"allows":[83],"fine-grained":[86],"mixing,":[87],"thereby":[88],"increasing":[89],"model's":[91],"ability":[92],"jointly":[94],"handle":[95],"multiple":[96],"tasks.We":[97],"evaluate":[98],"MoDE":[99],"on":[100],"Supernatural":[102],"Instructions":[103],"(SNI)":[104],"benchmark":[105],"consisting":[106],"set":[110],"700+":[112],"demonstrate":[115],"it":[117],"outperforms":[118],"state-of-the-art":[119],"parameter-efficient":[121],"(PEFT)":[123],"methods,":[124],"without":[125],"introducing":[126],"additional":[127],"parameters.Our":[128],"findings":[129],"contribute":[130],"deeper":[133],"understanding":[134],"parameter":[136],"efficiency":[137],"LLM":[140],"provide":[143],"practical":[145],"solution":[146],"deploying":[148],"highperforming,":[149],"lightweight":[150],"models.":[151]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-08-25T07:29:55.448023","created_date":"2025-10-10T00:00:00"}