{"id":"https://openalex.org/W4411542016","doi":"https://doi.org/10.1145/3715275.3732127","title":"Uncovering the Linguistic Roots of Bias: Insights and Mitigation in Large Language Models","display_name":"Uncovering the Linguistic Roots of Bias: Insights and Mitigation in Large Language Models","publication_year":2025,"publication_date":"2025-06-23","ids":{"openalex":"https://openalex.org/W4411542016","doi":"https://doi.org/10.1145/3715275.3732127"},"language":"en","primary_location":{"id":"doi:10.1145/3715275.3732127","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3715275.3732127","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3715275.3732127","source":null,"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 2025 ACM Conference on Fairness, Accountability, and Transparency","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3715275.3732127","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5037460608","display_name":"Lauren Benson","orcid":null},"institutions":[{"id":"https://openalex.org/I114662689","display_name":"Leidos (United States)","ror":"https://ror.org/012cvds63","country_code":"US","type":"company","lineage":["https://openalex.org/I114662689"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Lauren Benson","raw_affiliation_strings":["AI Accelerator, Leidos Inc, Reston, VA, USA"],"raw_orcid":"https://orcid.org/0009-0007-6876-0590","affiliations":[{"raw_affiliation_string":"AI Accelerator, Leidos Inc, Reston, VA, USA","institution_ids":["https://openalex.org/I114662689"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040805050","display_name":"Ahmet Okutan","orcid":"https://orcid.org/0000-0001-6664-515X"},"institutions":[{"id":"https://openalex.org/I114662689","display_name":"Leidos (United States)","ror":"https://ror.org/012cvds63","country_code":"US","type":"company","lineage":["https://openalex.org/I114662689"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ahmet Okutan","raw_affiliation_strings":["AI Accelerator, Leidos Inc, Reston, VA, USA"],"raw_orcid":"https://orcid.org/0000-0001-6664-515X","affiliations":[{"raw_affiliation_string":"AI Accelerator, Leidos Inc, Reston, VA, USA","institution_ids":["https://openalex.org/I114662689"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5055530849","display_name":"Roopa Vasan","orcid":null},"institutions":[{"id":"https://openalex.org/I114662689","display_name":"Leidos (United States)","ror":"https://ror.org/012cvds63","country_code":"US","type":"company","lineage":["https://openalex.org/I114662689"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Roopa Vasan","raw_affiliation_strings":["AI Accelerator, Leidos Inc, Reston, VA, USA"],"raw_orcid":"https://orcid.org/0009-0004-5158-6870","affiliations":[{"raw_affiliation_string":"AI Accelerator, Leidos Inc, Reston, VA, USA","institution_ids":["https://openalex.org/I114662689"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I114662689"],"apc_list":null,"apc_paid":null,"fwci":1.673,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.82084053,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"1916","last_page":"1926"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9456999897956848,"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/T10028","display_name":"Topic Modeling","score":0.9456999897956848,"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/T10883","display_name":"Ethics and Social Impacts of AI","score":0.9435999989509583,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6915512084960938},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.5466218590736389},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.49238407611846924},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.41803717613220215},{"id":"https://openalex.org/keywords/philosophy","display_name":"Philosophy","score":0.08461928367614746}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6915512084960938},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.5466218590736389},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.49238407611846924},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41803717613220215},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.08461928367614746}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3715275.3732127","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3715275.3732127","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3715275.3732127","source":null,"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 2025 ACM Conference on Fairness, Accountability, and Transparency","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3715275.3732127","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3715275.3732127","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3715275.3732127","source":null,"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 2025 ACM Conference on Fairness, Accountability, and Transparency","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4411542016.pdf","grobid_xml":"https://content.openalex.org/works/W4411542016.grobid-xml"},"referenced_works_count":16,"referenced_works":["https://openalex.org/W2559655401","https://openalex.org/W2750779823","https://openalex.org/W2963078909","https://openalex.org/W2972413484","https://openalex.org/W3015453090","https://openalex.org/W3034238904","https://openalex.org/W3105882417","https://openalex.org/W3176477796","https://openalex.org/W3181414820","https://openalex.org/W3207715300","https://openalex.org/W4287887133","https://openalex.org/W4293066130","https://openalex.org/W4389518334","https://openalex.org/W4391156274","https://openalex.org/W4396820380","https://openalex.org/W4401209579"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W3204019825"],"abstract_inverted_index":{"The":[0],"complexity,":[1],"diversity,":[2],"and":[3,16,26,36,55,59,74,122,154,159,188,195],"opacity":[4],"of":[5,93,119],"large":[6],"language":[7,80],"models":[8,95],"(LLMs)":[9],"present":[10],"significant":[11],"challenges":[12],"in":[13,23,167,180,190],"ensuring":[14],"fairness":[15,187],"trustworthiness.While":[17],"existing":[18],"research":[19],"has":[20],"made":[21],"strides":[22],"creating":[24],"interpretable":[25],"controllable":[27],"models,":[28],"there":[29],"remains":[30],"a":[31,45,97,136,164,172],"critical":[32],"need":[33],"for":[34,176],"generalizable":[35],"real-world-applicable":[37],"bias":[38,65,75,131,156],"mitigation":[39],"solutions.In":[40],"this":[41],"work,":[42],"we":[43,77],"propose":[44],"novel":[46],"methodology":[47],"that":[48,84,101,110],"bridges":[49],"the":[50,68,91,142,183],"gap":[51],"between":[52,70],"linguistic":[53,72],"capabilities":[54,114],"biased":[56,87,120],"behavior,":[57,121],"uncovering":[58],"leveraging":[60],"hidden":[61],"correlations":[62],"to":[63,129,141,185],"mitigate":[64,130],"holistically.By":[66],"analyzing":[67],"interplay":[69],"unrelated":[71],"features":[73],"benchmarks,":[76],"identify":[78],"natural":[79],"understanding":[81],"(NLU)":[82],"tasks":[83],"significantly":[85],"influence":[86],"behavior.This":[88],"insight":[89],"enables":[90],"creation":[92],"bias-mitigated":[94],"through":[96],"multi-task":[98],"learning":[99],"framework":[100],"augments":[102],"behavior":[103],"with":[104,116,162],"minimal":[105],"performance":[106],"trade-offs.Our":[107],"findings":[108],"reveal":[109],"(1)":[111],"specific":[112],"NLU":[113],"correlate":[115],"multiple":[117],"perspectives":[118],"(2)":[123],"these":[124],"relationships":[125],"can":[126],"be":[127],"applied":[128],"without":[132],"sacrificing":[133],"model":[134],"accuracy.In":[135],"real-world":[137],"domain-specific":[138],"application,":[139],"according":[140],"Aequitas":[143],"disparity":[144],"measures":[145],"using":[146],"client-provided":[147],"demographic":[148],"labels,":[149],"our":[150],"method":[151],"reduced":[152],"gender":[153],"age":[155],"by":[157],"14%":[158],"11%,":[160],"respectively,":[161],"only":[163],"0.03%":[165],"decrease":[166],"Micro-F1":[168],"score.This":[169],"approach":[170],"introduces":[171],"scalable,":[173],"generalized":[174],"pathway":[175],"addressing":[177],"systemic":[178],"biases":[179],"LLMs,":[181],"demonstrating":[182],"potential":[184],"improve":[186],"trustworthiness":[189],"AI":[191],"systems":[192],"across":[193],"diverse":[194],"sensitive":[196],"applications.":[197]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}