{"id":"https://openalex.org/W7162032519","doi":"https://doi.org/10.48550/arxiv.2605.21217","title":"Federated LoRA Fine-Tuning for LLMs via Collaborative Alignment","display_name":"Federated LoRA Fine-Tuning for LLMs via Collaborative Alignment","publication_year":2026,"publication_date":"2026-05-20","ids":{"openalex":"https://openalex.org/W7162032519","doi":"https://doi.org/10.48550/arxiv.2605.21217"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.21217","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.21217","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.21217","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136648299","display_name":"Shuaida He","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Shuaida","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100730193","display_name":"Liwen Chen","orcid":"https://orcid.org/0000-0002-9050-3829"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Liwen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136685043","display_name":"Long Feng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Feng, Long","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":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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.34769999980926514,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.34769999980926514,"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/T10028","display_name":"Topic Modeling","score":0.12759999930858612,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.06830000132322311,"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/subspace-topology","display_name":"Subspace topology","score":0.7534000277519226},{"id":"https://openalex.org/keywords/oracle","display_name":"Oracle","score":0.5110999941825867},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.4702000021934509},{"id":"https://openalex.org/keywords/adaptation","display_name":"Adaptation (eye)","score":0.392300009727478},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.38190001249313354},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.3190999925136566},{"id":"https://openalex.org/keywords/estimator","display_name":"Estimator","score":0.3089999854564667}],"concepts":[{"id":"https://openalex.org/C32834561","wikidata":"https://www.wikidata.org/wiki/Q660730","display_name":"Subspace topology","level":2,"score":0.7534000277519226},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7501999735832214},{"id":"https://openalex.org/C55166926","wikidata":"https://www.wikidata.org/wiki/Q2892946","display_name":"Oracle","level":2,"score":0.5110999941825867},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.4702000021934509},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.42739999294281006},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.40299999713897705},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.392300009727478},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.38190001249313354},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3497999906539917},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3393000066280365},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.33379998803138733},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.3190999925136566},{"id":"https://openalex.org/C185429906","wikidata":"https://www.wikidata.org/wiki/Q1130160","display_name":"Estimator","level":2,"score":0.3089999854564667},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.3061000108718872},{"id":"https://openalex.org/C137471899","wikidata":"https://www.wikidata.org/wiki/Q11381","display_name":"Local area network","level":2,"score":0.2847999930381775},{"id":"https://openalex.org/C96250715","wikidata":"https://www.wikidata.org/wiki/Q965330","display_name":"Estimation","level":2,"score":0.2741999924182892},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.2741999924182892},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.2694000005722046},{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.2630000114440918},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2599000036716461},{"id":"https://openalex.org/C2778334786","wikidata":"https://www.wikidata.org/wiki/Q1586270","display_name":"Variation (astronomy)","level":2,"score":0.25049999356269836}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.21217","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.21217","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.21217","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.21217","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Low-rank":[0,59],"adaptation":[1,90],"(LoRA)":[2],"has":[3],"emerged":[4],"as":[5],"a":[6,22,38,50,65,108,193,197],"powerful":[7],"tool":[8],"for":[9],"parameter-efficient":[10],"fine-tuning":[11,28,170,192,213],"of":[12,118,178,189],"large":[13],"language":[14],"models":[15],"(LLMs).":[16],"This":[17],"paper":[18],"studies":[19],"LoRA":[20,101,121,164],"under":[21,129,137],"federated":[23,216],"learning":[24],"setting,":[25],"enabling":[26],"collaborative":[27],"across":[29],"clients":[30,44,106],"while":[31,157],"preserving":[32,158],"parameter":[33],"efficiency.":[34],"We":[35,56,114,141],"focus":[36],"on":[37,71,196],"highly":[39],"heterogeneous":[40],"regime":[41],"in":[42,123],"which":[43],"share":[45],"only":[46,70],"partial":[47],"structure":[48],"and":[49,61,85,103,133,181,206,214],"substantial":[51],"subset":[52],"may":[53],"be":[54,92],"contaminated.":[55],"propose":[57],"Collaborative":[58],"Alignment":[60],"Identifiable":[62],"Recovery":[63],"(CLAIR),":[64],"contamination-aware":[66],"framework":[67],"that":[68],"relies":[69],"preliminary":[72,130],"local":[73,89,169,212],"estimators.":[74],"Its":[75],"formulation":[76],"applies":[77],"broadly,":[78],"from":[79,146],"linear":[80],"regression":[81],"to":[82],"neural":[83],"network":[84],"LLM":[86],"modules,":[87],"whenever":[88,171],"can":[91],"represented":[93],"by":[94,191],"matrix-valued":[95],"updates.":[96],"CLAIR":[97,147,190],"recovers":[98],"the":[99,119,124,144,162,176,187],"shared":[100,120,163],"subspace":[102,122,179],"detects":[104],"contaminated":[105],"via":[107],"structured":[109],"low-rank":[110],"plus":[111],"block-sparse":[112],"decomposition.":[113],"prove":[115],"exact":[116],"recovery":[117,128,136],"noiseless":[125],"case,":[126],"stable":[127],"estimation":[131,152,180],"error,":[132],"consistent":[134],"collaborative-set":[135],"mild":[138],"separation":[139],"conditions.":[140],"further":[142],"quantify":[143],"gain":[145,174],"refinement:":[148],"it":[149],"reduces":[150],"off-subspace":[151],"error":[153],"through":[154],"cross-client":[155],"averaging":[156],"client-specific":[159],"variation":[160],"within":[161],"subspace,":[165],"thus":[166],"improves":[167],"over":[168],"this":[172],"oracle":[173],"outweighs":[175],"costs":[177],"benign-client":[182,208],"heterogeneity.":[183],"Empirically,":[184],"we":[185],"demonstrate":[186],"benefits":[188],"Transformer":[194],"architecture":[195],"text-copying":[198],"task.":[199],"The":[200],"results":[201],"show":[202],"accurate":[203],"contamination":[204],"detection":[205],"improved":[207],"performance":[209],"compared":[210],"with":[211],"non-robust":[215],"averaging.":[217]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-05-22T00:00:00"}