{"id":"https://openalex.org/W7077911237","doi":"https://doi.org/10.48550/arxiv.2508.17337","title":"DropLoRA: Sparse Low-Rank Adaptation for Parameter-Efficient Fine-Tuning","display_name":"DropLoRA: Sparse Low-Rank Adaptation for Parameter-Efficient Fine-Tuning","publication_year":2025,"publication_date":"2025-08-24","ids":{"openalex":"https://openalex.org/W7077911237","doi":"https://doi.org/10.48550/arxiv.2508.17337"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2508.17337","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.17337","pdf_url":"https://arxiv.org/pdf/2508.17337","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,"listed_in":[],"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2508.17337","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Zhang, Haojie","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Zhang, Haojie","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":true,"primary_topic":{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.8179000020027161,"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":0.8179000020027161,"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":0.1469999998807907,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.01850000023841858,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/lora","display_name":"LoRa","score":0.8859999775886536},{"id":"https://openalex.org/keywords/parameter-efficient-fine-tuning","display_name":"parameter-efficient fine-tuning","score":0.9190000295639038},{"id":"https://openalex.org/keywords/low-rank-adaptation","display_name":"low-rank adaptation","score":0.8679999709129333},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.8830000162124634},{"id":"https://openalex.org/keywords/llama","display_name":"llamas","score":0.890999972820282},{"id":"https://openalex.org/keywords/code-generation","display_name":"code generation","score":0.5400000214576721},{"id":"https://openalex.org/keywords/instruction-following","display_name":"instruction following","score":0.5690000057220459},{"id":"https://openalex.org/keywords/mathematical-reasoning","display_name":"mathematical reasoning","score":0.8360000252723694},{"id":"https://openalex.org/keywords/commonsense-reasoning","display_name":"commonsense reasoning","score":0.968999981880188}],"concepts":[{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.7875000238418579},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7529000043869019},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.6693999767303467},{"id":"https://openalex.org/C164226766","wikidata":"https://www.wikidata.org/wiki/Q7293202","display_name":"Rank (graph theory)","level":2,"score":0.6425999999046326},{"id":"https://openalex.org/C32834561","wikidata":"https://www.wikidata.org/wiki/Q660730","display_name":"Subspace topology","level":2,"score":0.625},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5432999730110168},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5232999920845032},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.5},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4650999903678894},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.3926999866962433},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.3093999922275543},{"id":"https://openalex.org/C143724316","wikidata":"https://www.wikidata.org/wiki/Q312468","display_name":"Series (stratigraphy)","level":2,"score":0.30160000920295715},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.2883000075817108},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.2863999903202057},{"id":"https://openalex.org/C19768560","wikidata":"https://www.wikidata.org/wiki/Q320727","display_name":"Dependency (UML)","level":2,"score":0.2752000093460083},{"id":"https://openalex.org/C86037889","wikidata":"https://www.wikidata.org/wiki/Q4330127","display_name":"Learning to rank","level":3,"score":0.27250000834465027},{"id":"https://openalex.org/C148220186","wikidata":"https://www.wikidata.org/wiki/Q7111912","display_name":"Outcome (game theory)","level":2,"score":0.26750001311302185},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.25999999046325684},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.25130000710487366}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2508.17337","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.17337","pdf_url":"https://arxiv.org/pdf/2508.17337","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,"listed_in":[],"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2508.17337","doi":"https://doi.org/10.48550/arxiv.2508.17337","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2508.17337","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,"listed_in":[],"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2508.17337","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.17337","pdf_url":"https://arxiv.org/pdf/2508.17337","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,"listed_in":[],"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"LoRA-based":[0],"large":[1,140],"model":[2,15,142],"parameter-efficient":[3],"fine-tuning":[4,131],"(PEFT)":[5],"methods":[6,54],"use":[7],"low-rank":[8,23,60,71],"de-":[9],"composition":[10],"to":[11,14,19,27,57,75,89],"approximate":[12],"updates":[13,24],"parameters.":[16],"However,":[17],"compared":[18],"full-":[20],"parameter":[21],"fine-tuning,":[22],"often":[25],"lead":[26],"a":[28,40,65,99,136],"performance":[29,111],"gap":[30],"in":[31,73,130],"downstream":[32],"tasks.":[33],"To":[34],"address":[35],"this,":[36],"we":[37],"introduce":[38],"DropLoRA,":[39],"novel":[41],"pruning-based":[42],"approach":[43],"that":[44,55,125],"focuses":[45],"on":[46],"pruning":[47,66],"the":[48,59,69,91,105,132],"rank":[49,84],"dimension.":[50],"Unlike":[51],"conven-":[52],"tional":[53],"attempt":[56],"overcome":[58,90],"bottleneck,":[61],"DropLoRA":[62,88,108,126],"innovatively":[63],"integrates":[64],"module":[67],"between":[68],"two":[70],"matrices":[72],"LoRA":[74,129],"simulate":[76],"dy-":[77],"namic":[78],"subspace":[79,85],"learning.":[80],"This":[81],"dynamic":[82],"low-":[83],"learning":[86,106],"allows":[87],"limitations":[92],"of":[93,139],"traditional":[94],"LoRA,":[95],"which":[96],"operates":[97],"within":[98],"static":[100],"subspace.":[101],"By":[102],"continuously":[103],"adapting":[104],"subspace,":[107],"significantly":[109],"boosts":[110],"without":[112],"incurring":[113],"additional":[114],"training":[115],"or":[116],"infer-":[117],"ence":[118],"costs.":[119],"Our":[120,156],"experimental":[121],"results":[122],"demon-":[123],"strate":[124],"consistently":[127],"outperforms":[128],"LLaMA":[133],"series":[134],"across":[135],"wide":[137],"range":[138],"language":[141],"gener-":[143],"ation":[144],"tasks,":[145],"including":[146],"commonsense":[147],"reason-":[148],"ing,":[149],"mathematical":[150],"reasoning,":[151],"code":[152,157],"generation,":[153],"and":[154],"instruction-following.":[155],"is":[158],"avail-":[159],"able":[160],"at":[161],"https://github.com/TayeeChang/DropLoRA.":[162]},"counts_by_year":[],"updated_date":"2026-09-25T08:18:29.962427","created_date":"2025-10-10T00:00:00"}