{"id":"https://openalex.org/W4285273243","doi":"https://doi.org/10.18653/v1/2022.ltedi-1.8","title":"Debiasing Pre-Trained Language Models via Efficient Fine-Tuning","display_name":"Debiasing Pre-Trained Language Models via Efficient Fine-Tuning","publication_year":2022,"publication_date":"2022-01-01","ids":{"openalex":"https://openalex.org/W4285273243","doi":"https://doi.org/10.18653/v1/2022.ltedi-1.8"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2022.ltedi-1.8","doi":"https://doi.org/10.18653/v1/2022.ltedi-1.8","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2022.ltedi-1.8","pdf_url":"https://aclanthology.org/2022.ltedi-1.8.pdf","source":{"id":"https://openalex.org/S7407107002","display_name":"Workshop on Language Technology for Equality, Diversity, Inclusion (LT-EDI)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Second Workshop on Language Technology for Equality, Diversity and Inclusion","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://aclanthology.org/2022.ltedi-1.8.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5062409164","display_name":"Michael Gira","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I135310074","display_name":"University of Wisconsin\u2013Madison","ror":"https://ror.org/01y2jtd41","country_code":"US","type":"education","lineage":["https://openalex.org/I135310074"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Michael Gira","raw_affiliation_strings":["University of Wisconsin-Madison"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Wisconsin-Madison","institution_ids":["https://openalex.org/I135310074"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050795462","display_name":"Ruisu Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I135310074","display_name":"University of Wisconsin\u2013Madison","ror":"https://ror.org/01y2jtd41","country_code":"US","type":"education","lineage":["https://openalex.org/I135310074"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ruisu Zhang","raw_affiliation_strings":["University of Wisconsin-Madison"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Wisconsin-Madison","institution_ids":["https://openalex.org/I135310074"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101899300","display_name":"Kangwook Lee","orcid":"https://orcid.org/0000-0002-3360-9678","observed_orcids":["https://orcid.org/0000-0002-3360-9678"]},"institutions":[{"id":"https://openalex.org/I135310074","display_name":"University of Wisconsin\u2013Madison","ror":"https://ror.org/01y2jtd41","country_code":"US","type":"education","lineage":["https://openalex.org/I135310074"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kangwook Lee","raw_affiliation_strings":["University of Wisconsin-Madison"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Wisconsin-Madison","institution_ids":["https://openalex.org/I135310074"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I135310074"],"apc_list":null,"apc_paid":null,"fwci":7.2047,"has_fulltext":true,"cited_by_count":32,"citation_normalized_percentile":{"value":0.96819507,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":93,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"59","last_page":"69"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10883","display_name":"Ethics and Social Impacts of AI","score":0.8705999851226807,"subfield":{"id":"https://openalex.org/subfields/3311","display_name":"Safety Research"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T10883","display_name":"Ethics and Social Impacts of AI","score":0.8705999851226807,"subfield":{"id":"https://openalex.org/subfields/3311","display_name":"Safety Research"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.06859999895095825,"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/T12262","display_name":"Hate Speech and Cyberbullying Detection","score":0.020099999383091927,"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/bias-mitigation","display_name":"bias mitigation","score":0.8709999918937683},{"id":"https://openalex.org/keywords/gender-bias","display_name":"gender bias","score":0.9139999747276306},{"id":"https://openalex.org/keywords/pre-trained-language-models","display_name":"pre-trained language models","score":0.9089999794960022},{"id":"https://openalex.org/keywords/efficient-fine-tuning","display_name":"efficient fine-tuning","score":0.9039999842643738},{"id":"https://openalex.org/keywords/gpt-2","display_name":"GPT-2","score":0.8600000143051147},{"id":"https://openalex.org/keywords/catastrophic-forgetting","display_name":"catastrophic forgetting","score":0.8420000076293945},{"id":"https://openalex.org/keywords/transformer","display_name":"transformer models","score":0.34200000762939453}],"concepts":[{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.7805112600326538},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7582405805587769},{"id":"https://openalex.org/C2779458634","wikidata":"https://www.wikidata.org/wiki/Q24963715","display_name":"Debiasing","level":2,"score":0.7035970091819763},{"id":"https://openalex.org/C7149132","wikidata":"https://www.wikidata.org/wiki/Q1377840","display_name":"Forgetting","level":2,"score":0.661439061164856},{"id":"https://openalex.org/C2780586970","wikidata":"https://www.wikidata.org/wiki/Q1357284","display_name":"Popularity","level":2,"score":0.6167502999305725},{"id":"https://openalex.org/C2776050585","wikidata":"https://www.wikidata.org/wiki/Q7439360","display_name":"Scrutiny","level":2,"score":0.510057270526886},{"id":"https://openalex.org/C125209513","wikidata":"https://www.wikidata.org/wiki/Q4037520","display_name":"Doors","level":2,"score":0.4661461412906647},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.46040236949920654},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.45170533657073975},{"id":"https://openalex.org/C132459708","wikidata":"https://www.wikidata.org/wiki/Q744069","display_name":"Extrapolation","level":2,"score":0.4237526059150696},{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.4207378923892975},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3489382266998291},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.11653348803520203},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.08860602974891663},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.08728548884391785},{"id":"https://openalex.org/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","level":1,"score":0.0},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C77805123","wikidata":"https://www.wikidata.org/wiki/Q161272","display_name":"Social psychology","level":1,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2022.ltedi-1.8","doi":"https://doi.org/10.18653/v1/2022.ltedi-1.8","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2022.ltedi-1.8","pdf_url":"https://aclanthology.org/2022.ltedi-1.8.pdf","source":{"id":"https://openalex.org/S7407107002","display_name":"Workshop on Language Technology for Equality, Diversity, Inclusion (LT-EDI)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Second Workshop on Language Technology for Equality, Diversity and Inclusion","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2022.ltedi-1.8","doi":"https://doi.org/10.18653/v1/2022.ltedi-1.8","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2022.ltedi-1.8","pdf_url":"https://aclanthology.org/2022.ltedi-1.8.pdf","source":{"id":"https://openalex.org/S7407107002","display_name":"Workshop on Language Technology for Equality, Diversity, Inclusion (LT-EDI)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Second Workshop on Language Technology for Equality, Diversity and Inclusion","raw_type":"proceedings-article"},"sustainable_development_goals":[{"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10","score":0.7578999996185303},{"display_name":"Gender equality","id":"https://metadata.un.org/sdg/5","score":0.6025999784469604}],"study_designs":[],"awards":[{"id":"https://openalex.org/G4649966841","display_name":"MLWiNS: Distributed Learning for the Nomadic Edge","funder_award_id":"2003129","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320306555","display_name":"Wisconsin Alumni Research Foundation","ror":"https://ror.org/00hwxbz16"},{"id":"https://openalex.org/F4320309434","display_name":"University of Wisconsin-Madison","ror":"https://ror.org/01y2jtd41"},{"id":"https://openalex.org/F4320337368","display_name":"Division of Graduate Education","ror":"https://ror.org/00whkrf32"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4285273243.pdf","grobid_xml":"https://content.openalex.org/works/W4285273243.grobid-xml"},"referenced_works_count":45,"referenced_works":["https://openalex.org/W1614298861","https://openalex.org/W2117539524","https://openalex.org/W2140534852","https://openalex.org/W2217426128","https://openalex.org/W2250539671","https://openalex.org/W2253429366","https://openalex.org/W2279176662","https://openalex.org/W2559153497","https://openalex.org/W2560647685","https://openalex.org/W2591924527","https://openalex.org/W2622808887","https://openalex.org/W2626778328","https://openalex.org/W2770618123","https://openalex.org/W2787560479","https://openalex.org/W2790025105","https://openalex.org/W2889624842","https://openalex.org/W2893425640","https://openalex.org/W2922862135","https://openalex.org/W2926555354","https://openalex.org/W2946280906","https://openalex.org/W2950018712","https://openalex.org/W2952992882","https://openalex.org/W2962059918","https://openalex.org/W2963290659","https://openalex.org/W2963341956","https://openalex.org/W2963526187","https://openalex.org/W2965373594","https://openalex.org/W2969896603","https://openalex.org/W2971307358","https://openalex.org/W2972413484","https://openalex.org/W2975059944","https://openalex.org/W3012624518","https://openalex.org/W3019416653","https://openalex.org/W3030163527","https://openalex.org/W3041133507","https://openalex.org/W3089430725","https://openalex.org/W3094502228","https://openalex.org/W3094997972","https://openalex.org/W3108091790","https://openalex.org/W3113184484","https://openalex.org/W3133702157","https://openalex.org/W3134307371","https://openalex.org/W3135367836","https://openalex.org/W3175060421","https://openalex.org/W4386290290"],"related_works":["https://openalex.org/W4379958573","https://openalex.org/W4385571213","https://openalex.org/W4284899401","https://openalex.org/W4385571122","https://openalex.org/W3154654049","https://openalex.org/W4402683076","https://openalex.org/W7126376559","https://openalex.org/W4399111298","https://openalex.org/W7127434609","https://openalex.org/W4285192297"],"abstract_inverted_index":{"An":[0],"explosion":[1],"in":[2,46,81,107],"the":[3,18,62],"popularity":[4],"of":[5,91],"transformerbased":[6],"language":[7,26,37,84,109],"models":[8,38,110],"(such":[9],"as":[10],"GPT-3,":[11],"BERT,":[12],"RoBERTa,":[13],"and":[14,30],"ALBERT)":[15],"has":[16,51],"opened":[17],"doors":[19],"to":[20,60,72,77,104],"new":[21],"machine":[22],"learning":[23],"applications":[24],"involving":[25],"modeling,":[27],"text":[28],"generation,":[29],"more.However,":[31],"recent":[32],"scrutiny":[33],"reveals":[34],"that":[35,98],"these":[36],"contain":[39],"inherent":[40],"biases":[41],"towards":[42],"certain":[43],"demographics":[44],"reflected":[45],"their":[47],"training":[48],"data.While":[49],"research":[50],"tried":[52],"mitigating":[53],"this":[54,99],"problem,":[55],"existing":[56],"approaches":[57],"either":[58],"fail":[59],"remove":[61],"bias":[63,80],"completely,":[64],"degrade":[65],"performance":[66],"(\"catastrophic":[67],"forgetting\"),":[68],"or":[69],"are":[70],"costly":[71],"execute.This":[73],"work":[74],"examines":[75],"how":[76],"reduce":[78,105],"gender":[79],"a":[82,101],"GPT-2":[83],"model":[85],"by":[86],"fine-tuning":[87],"less":[88],"than":[89],"1%":[90],"its":[92],"parameters.Through":[93],"quantitative":[94],"benchmarks,":[95],"we":[96],"show":[97],"is":[100],"viable":[102],"way":[103],"prejudice":[106],"pre-trained":[108],"while":[111],"remaining":[112],"cost-effective":[113],"at":[114],"scale.":[115]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":14},{"year":2023,"cited_by_count":10},{"year":2022,"cited_by_count":2}],"updated_date":"2026-10-29T09:56:32.521200","created_date":"2022-07-14T00:00:00"}