{"id":"https://openalex.org/W4416894531","doi":"https://doi.org/10.3389/fdata.2025.1677331","title":"Parameter-efficient fine-tuning for low-resource text classification: a comparative study of LoRA, IA3, and ReFT","display_name":"Parameter-efficient fine-tuning for low-resource text classification: a comparative study of LoRA, IA3, and ReFT","publication_year":2025,"publication_date":"2025-12-02","ids":{"openalex":"https://openalex.org/W4416894531","doi":"https://doi.org/10.3389/fdata.2025.1677331","pmid":"https://pubmed.ncbi.nlm.nih.gov/41409043"},"language":"en","primary_location":{"id":"doi:10.3389/fdata.2025.1677331","is_oa":true,"landing_page_url":"https://doi.org/10.3389/fdata.2025.1677331","pdf_url":"https://public-pages-files-2025.frontiersin.org/journals/big-data/articles/10.3389/fdata.2025.1677331/pdf","source":{"id":"https://openalex.org/S4210201220","display_name":"Frontiers in Big Data","issn_l":"2624-909X","issn":["2624-909X"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320527","host_organization_name":"Frontiers Media","host_organization_lineage":["https://openalex.org/P4310320527"],"host_organization_lineage_names":["Frontiers Media"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Big Data","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","doaj","pubmed"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://public-pages-files-2025.frontiersin.org/journals/big-data/articles/10.3389/fdata.2025.1677331/pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5115464003","display_name":"Steve Nwaiwu","orcid":null},"institutions":[{"id":"https://openalex.org/I10245363","display_name":"Rajamangala University of Technology","ror":"https://ror.org/051qqcg15","country_code":"TH","type":"education","lineage":["https://openalex.org/I10245363"]}],"countries":["TH"],"is_corresponding":true,"raw_author_name":"Steve Nwaiwu","raw_affiliation_strings":["Data and Information Science, Faculty of Science and Technology, Rajamangala University of Technology, Pathum Thani, Thailand"],"affiliations":[{"raw_affiliation_string":"Data and Information Science, Faculty of Science and Technology, Rajamangala University of Technology, Pathum Thani, Thailand","institution_ids":["https://openalex.org/I10245363"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5115464003"],"corresponding_institution_ids":["https://openalex.org/I10245363"],"apc_list":{"value":1150,"currency":"USD","value_usd":1150},"apc_paid":{"value":1150,"currency":"USD","value_usd":1150},"fwci":2.6124,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.9270174,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":99},"biblio":{"volume":"8","issue":null,"first_page":"1677331","last_page":"1677331"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12535","display_name":"Machine Learning and Data Classification","score":0.2125999927520752,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.2125999927520752,"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/T10664","display_name":"Sentiment Analysis and Opinion Mining","score":0.11249999701976776,"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":0.0763000026345253,"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/overfitting","display_name":"Overfitting","score":0.6682000160217285},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5163999795913696},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.46059998869895935},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.4323999881744385},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.4146000146865845},{"id":"https://openalex.org/keywords/natural-language-understanding","display_name":"Natural language understanding","score":0.361299991607666},{"id":"https://openalex.org/keywords/question-answering","display_name":"Question answering","score":0.35350000858306885},{"id":"https://openalex.org/keywords/baseline","display_name":"Baseline (sea)","score":0.33149999380111694}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7289000153541565},{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.6682000160217285},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5846999883651733},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5400000214576721},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5163999795913696},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.46059998869895935},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.4323999881744385},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.4146000146865845},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.39590001106262207},{"id":"https://openalex.org/C2779439875","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Natural language understanding","level":3,"score":0.361299991607666},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.35350000858306885},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.33149999380111694},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.30649998784065247},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.2924000024795532},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.28700000047683716},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.2827000021934509},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.28130000829696655},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.2685000002384186},{"id":"https://openalex.org/C155092808","wikidata":"https://www.wikidata.org/wiki/Q182557","display_name":"Computational linguistics","level":2,"score":0.2660999894142151},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.260699987411499},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.2574000060558319},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.25200000405311584}],"mesh":[],"locations_count":5,"locations":[{"id":"doi:10.3389/fdata.2025.1677331","is_oa":true,"landing_page_url":"https://doi.org/10.3389/fdata.2025.1677331","pdf_url":"https://public-pages-files-2025.frontiersin.org/journals/big-data/articles/10.3389/fdata.2025.1677331/pdf","source":{"id":"https://openalex.org/S4210201220","display_name":"Frontiers in Big Data","issn_l":"2624-909X","issn":["2624-909X"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320527","host_organization_name":"Frontiers Media","host_organization_lineage":["https://openalex.org/P4310320527"],"host_organization_lineage_names":["Frontiers Media"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Big Data","raw_type":"journal-article"},{"id":"pmid:41409043","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/41409043","pdf_url":null,"source":{"id":"https://openalex.org/S4306525036","display_name":"PubMed","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in big data","raw_type":null},{"id":"pmh:oai:doaj.org/article:fc8f888f07f7415db1e65b824534cbe4","is_oa":true,"landing_page_url":"https://doaj.org/article/fc8f888f07f7415db1e65b824534cbe4","pdf_url":null,"source":{"id":"https://openalex.org/S112646816","display_name":"SHILAP Revista de lepidopterolog\u00eda","issn_l":"0300-5267","issn":["0300-5267","2340-4078"],"is_oa":true,"is_in_doaj":true,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Frontiers in Big Data, Vol 8 (2025)","raw_type":"article"},{"id":"pmh:oai:europepmc.org:11514340","is_oa":true,"landing_page_url":"https://www.ncbi.nlm.nih.gov/pmc/articles/12705377","pdf_url":null,"source":{"id":"https://openalex.org/S4306400806","display_name":"Europe PMC (PubMed Central)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1303153112","host_organization_name":"European Bioinformatics Institute","host_organization_lineage":["https://openalex.org/I1303153112"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Text"},{"id":"pmh:oai:pubmedcentral.nih.gov:12705377","is_oa":true,"landing_page_url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC12705377/","pdf_url":null,"source":{"id":"https://openalex.org/S2764455111","display_name":"PubMed Central","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Front Big Data","raw_type":"Text"}],"best_oa_location":{"id":"doi:10.3389/fdata.2025.1677331","is_oa":true,"landing_page_url":"https://doi.org/10.3389/fdata.2025.1677331","pdf_url":"https://public-pages-files-2025.frontiersin.org/journals/big-data/articles/10.3389/fdata.2025.1677331/pdf","source":{"id":"https://openalex.org/S4210201220","display_name":"Frontiers in Big Data","issn_l":"2624-909X","issn":["2624-909X"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320527","host_organization_name":"Frontiers Media","host_organization_lineage":["https://openalex.org/P4310320527"],"host_organization_lineage_names":["Frontiers Media"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Big Data","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320326420","display_name":"Rajamangala University of Technology Thanyaburi","ror":"https://ror.org/051qqcg15"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416894531.pdf","grobid_xml":"https://content.openalex.org/works/W4416894531.grobid-xml"},"referenced_works_count":2,"referenced_works":["https://openalex.org/W3176828726","https://openalex.org/W4410061981"],"related_works":[],"abstract_inverted_index":{"The":[0,139],"successful":[1],"application":[2],"of":[3,23,63,79,110,169,177],"large-scale":[4],"transformer":[5],"models":[6],"in":[7,31,205],"Natural":[8],"Language":[9],"Processing":[10],"(NLP)":[11],"is":[12,28,70,192],"often":[13],"hindered":[14],"by":[15,89],"the":[16,111,118,149,154,178,183,216],"substantial":[17],"computational":[18],"cost":[19],"and":[20,42,91,96,114,135,233],"data":[21],"requirements":[22],"full":[24,126],"fine-tuning.":[25],"This":[26,72,187],"challenge":[27],"particularly":[29],"acute":[30],"low-resource":[32,68,108],"settings,":[33],"where":[34],"standard":[35],"fine-tuning":[36,127],"can":[37],"lead":[38],"to":[39,214],"catastrophic":[40],"overfitting":[41],"model":[43,106],"collapse.":[44],"To":[45],"address":[46],"this,":[47],"Parameter-Efficient":[48],"Fine-Tuning":[49,99],"(PEFT)":[50],"methods":[51,123,146],"have":[52],"emerged":[53],"as":[54,182],"a":[55,59,75,97,103,125,199,211,235],"promising":[56],"solution.":[57],"However,":[58],"direct":[60],"comparative":[61],"analysis":[62],"their":[64],"trade-offs":[65],"under":[66,223],"unified":[67],"conditions":[69],"lacking.":[71],"study":[73,227],"provides":[74],"rigorous":[76],"empirical":[77],"evaluation":[78],"three":[80],"prominent":[81],"PEFT":[82,145,191,240],"methods:":[83],"Low-Rank":[84],"Adaptation":[85],"(LoRA),":[86],"Infused":[87],"Adapter":[88],"Inhibiting":[90],"Amplifying":[92],"Inner":[93],"Activations":[94],"(IA<sup>3</sup>),":[95],"Representation":[98],"(ReFT)":[100],"strategy.":[101],"Using":[102],"DistilBERT":[104],"base":[105],"on":[107,159],"versions":[109],"AG":[112],"News":[113],"Amazon":[115,160],"Reviews":[116],"datasets,":[117],"present":[119],"work":[120],"compares":[121],"these":[122,221],"against":[124],"baseline":[128],"across":[129],"accuracy,":[130],"F1":[131,156,171],"score,":[132],"trainable":[133],"parameters,":[134,179],"GPU":[136],"memory":[137],"usage.":[138],"findings":[140],"reveal":[141],"that":[142,190],"while":[143,173],"all":[144],"dramatically":[147],"outperform":[148],"baseline,":[150],"LoRA":[151],"consistently":[152],"achieves":[153],"highest":[155],"scores":[157],"(0.909":[158],"Reviews).":[161],"Critically,":[162],"ReFT":[163],"delivers":[164],"nearly":[165],"identical":[166],"performance":[167],"(~98%":[168],"LoRA's":[170],"score)":[172],"training":[174],"only":[175],"~3%":[176],"establishing":[180],"it":[181],"most":[184],"efficient":[185],"method.":[186],"research":[188,232],"demonstrates":[189],"not":[193],"merely":[194],"an":[195],"efficiency":[196],"optimization,":[197],"but":[198],"necessary":[200],"tool":[201],"for":[202,238],"robust":[203],"generalization":[204],"data-scarce":[206],"environments,":[207],"providing":[208],"practitioners":[209],"with":[210],"clear":[212],"guide":[213],"navigate":[215],"performance-efficiency":[217],"trade-off.":[218],"By":[219],"unifying":[220],"evaluations":[222],"controlled":[224],"conditions,":[225],"this":[226],"advances":[228],"beyond":[229],"fragmented":[230],"prior":[231],"offers":[234],"systematic":[236],"framework":[237],"selecting":[239],"strategies.":[241]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2025-12-02T00:00:00"}