{"id":"https://openalex.org/W7148273218","doi":"https://doi.org/10.48550/arxiv.2604.00050","title":"Task-Centric Personalized Federated Fine-Tuning of Language Models","display_name":"Task-Centric Personalized Federated Fine-Tuning of Language Models","publication_year":2026,"publication_date":"2026-03-30","ids":{"openalex":"https://openalex.org/W7148273218","doi":"https://doi.org/10.48550/arxiv.2604.00050"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.00050","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.00050","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.00050","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5092278916","display_name":"Gabriel U. Talasso","orcid":"https://orcid.org/0009-0007-6422-9735"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Talasso, Gabriel U.","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132799584","display_name":"Meghdad Kurmanji","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kurmanji, Meghdad","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132807497","display_name":"Allan M. de Souza","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"de Souza, Allan M.","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132826837","display_name":"Nicholas D. Lane","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lane, Nicholas D.","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5132817781","display_name":"Leandro A. Villas","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Villas, Leandro A.","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5092278916"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.39750000834465027,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.39750000834465027,"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/T11719","display_name":"Data Quality and Management","score":0.1216999962925911,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.06920000165700912,"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/cluster-analysis","display_name":"Cluster analysis","score":0.5995000004768372},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.5311999917030334},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5260999798774719},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.5242999792098999},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.49300000071525574},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.4562000036239624},{"id":"https://openalex.org/keywords/adapter","display_name":"Adapter (computing)","score":0.45320001244544983},{"id":"https://openalex.org/keywords/data-modeling","display_name":"Data modeling","score":0.40049999952316284}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8450999855995178},{"id":"https://openalex.org/C73555534","wikidata":"https://www.wikidata.org/wiki/Q622825","display_name":"Cluster analysis","level":2,"score":0.5995000004768372},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.5311999917030334},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5260999798774719},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5242999792098999},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.499099999666214},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.49300000071525574},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4749999940395355},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.4562000036239624},{"id":"https://openalex.org/C177284502","wikidata":"https://www.wikidata.org/wiki/Q1005390","display_name":"Adapter (computing)","level":2,"score":0.45320001244544983},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.40049999952316284},{"id":"https://openalex.org/C2775896111","wikidata":"https://www.wikidata.org/wiki/Q642560","display_name":"Router","level":2,"score":0.37299999594688416},{"id":"https://openalex.org/C28006648","wikidata":"https://www.wikidata.org/wiki/Q6934509","display_name":"Multi-task learning","level":3,"score":0.35929998755455017},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3418999910354614},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.3230000138282776},{"id":"https://openalex.org/C2779585090","wikidata":"https://www.wikidata.org/wiki/Q3457762","display_name":"Resilience (materials science)","level":2,"score":0.30059999227523804},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.2962000072002411},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.29409998655319214},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.29010000824928284},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2888999879360199},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.2802000045776367},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.27410000562667847},{"id":"https://openalex.org/C72634772","wikidata":"https://www.wikidata.org/wiki/Q386824","display_name":"Data integration","level":2,"score":0.26260000467300415},{"id":"https://openalex.org/C16910744","wikidata":"https://www.wikidata.org/wiki/Q7705759","display_name":"Test data","level":2,"score":0.2533999979496002},{"id":"https://openalex.org/C2992525071","wikidata":"https://www.wikidata.org/wiki/Q50818671","display_name":"Federated learning","level":2,"score":0.2524999976158142}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.00050","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.00050","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.00050","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.00050","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Federated":[0],"Learning":[1],"(FL)":[2],"has":[3],"emerged":[4],"as":[5],"a":[6,89,113,156,199],"promising":[7],"technique":[8],"for":[9,48,120,125],"training":[10],"language":[11],"models":[12,23,46,119,133],"on":[13,25,73,187],"distributed":[14],"and":[15,83,155,219],"private":[16],"datasets":[17],"of":[18,33],"diverse":[19],"tasks.":[20,144],"However,":[21],"aggregating":[22],"trained":[24],"heterogeneous":[26],"tasks":[27,86,217],"often":[28],"degrades":[29],"the":[30,183,188],"overall":[31],"performance":[32],"individual":[34],"clients.":[35],"To":[36,105],"address":[37],"this":[38],"issue,":[39],"Personalized":[40],"FL":[41],"(pFL)":[42],"aims":[43],"to":[44,131,139,166,182,212,221],"create":[45],"tailored":[47],"each":[49,100,121,126],"client's":[50,91],"data":[51,81,92,153],"distribution.":[52],"Although":[53],"these":[54,107,207],"approaches":[55,197],"improve":[56],"local":[57,103,146],"performance,":[58],"they":[59],"usually":[60],"lack":[61],"robustness":[62],"in":[63,79,206],"two":[64,108,136],"aspects:":[65],"(i)":[66],"generalization:":[67],"when":[68,88],"clients":[69,165],"must":[70],"make":[71],"predictions":[72],"unseen":[74],"tasks,":[75],"or":[76],"face":[77],"changes":[78],"their":[80],"distributions,":[82],"(ii)":[84],"intra-client":[85],"interference:":[87],"single":[90],"contains":[93],"multiple":[94],"distributions":[95],"that":[96,116,148,159,178],"may":[97],"interfere":[98],"with":[99,142,151,195],"other":[101],"during":[102],"training.":[104],"tackle":[106],"challenges,":[109],"we":[110,172],"propose":[111,173],"FedRouter,":[112],"clustering-based":[114],"pFL":[115],"builds":[117],"specialized":[118],"task":[122,152],"rather":[123],"than":[124],"client.":[127],"FedRouter":[128,202],"uses":[129],"adapters":[130,141,150,162],"personalize":[132],"by":[134],"employing":[135],"clustering":[137,147],"mechanisms":[138],"associate":[140,149],"specific":[143],"A":[145],"samples":[154,181],"global":[157],"one":[158],"associates":[160],"similar":[161],"from":[163],"different":[164],"construct":[167],"task-centric":[168],"personalized":[169],"models.":[170],"Additionally,":[171],"an":[174],"evaluation":[175],"router":[176],"mechanism":[177],"routes":[179],"test":[180],"best":[184],"adapter":[185],"based":[186],"created":[189],"clusters.":[190],"Experiments":[191],"comparing":[192],"our":[193],"method":[194],"existing":[196],"across":[198],"multitask":[200],"dataset,":[201],"demonstrate":[203],"strong":[204],"resilience":[205],"challenging":[208],"scenarios":[209],"performing":[210],"up":[211,220],"6.1%":[213],"relatively":[214],"better":[215],"under":[216,225],"interference":[218],"136%":[222],"relative":[223],"improvement":[224],"generalization":[226],"evaluation.":[227]},"counts_by_year":[],"updated_date":"2026-04-03T16:44:17.987007","created_date":"2026-04-03T00:00:00"}