{"id":"https://openalex.org/W3198805641","doi":"https://doi.org/10.48550/arxiv.2109.03300","title":"Hi, my name is Martha: Using names to measure and mitigate bias in generative dialogue models","display_name":"Hi, my name is Martha: Using names to measure and mitigate bias in generative dialogue models","publication_year":2021,"publication_date":"2021-09-07","ids":{"openalex":"https://openalex.org/W3198805641","doi":"https://doi.org/10.48550/arxiv.2109.03300","mag":"3198805641"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2109.03300","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2109.03300","pdf_url":"https://arxiv.org/pdf/2109.03300","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/2109.03300","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Smith, Eric Michael","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210111288","display_name":"Meta (United Kingdom)","ror":"https://ror.org/020ye1821","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210111288","https://openalex.org/I4210114444"]},{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Smith, Eric Michael","raw_affiliation_strings":["Facebook"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Facebook","institution_ids":["https://openalex.org/I4210111288","https://openalex.org/I4210114444"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5062696000","display_name":"Adina Williams","orcid":"https://orcid.org/0000-0001-5281-3343","observed_orcids":["https://orcid.org/0000-0001-5281-3343"]},"institutions":[{"id":"https://openalex.org/I4210111288","display_name":"Meta (United Kingdom)","ror":"https://ror.org/020ye1821","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210111288","https://openalex.org/I4210114444"]},{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Williams, Adina","raw_affiliation_strings":["Facebook"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Facebook","institution_ids":["https://openalex.org/I4210111288","https://openalex.org/I4210114444"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":13,"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/T10883","display_name":"Ethics and Social Impacts of AI","score":0.9818999767303467,"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":0.9818999767303467,"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/T12031","display_name":"Speech and dialogue systems","score":0.004699999932199717,"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":0.0032999999821186066,"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/gender-bias","display_name":"gender bias","score":0.5870000123977661},{"id":"https://openalex.org/keywords/racial-bias","display_name":"racial bias","score":0.7379999756813049},{"id":"https://openalex.org/keywords/controlled-generation","display_name":"controlled generation","score":0.7639999985694885},{"id":"https://openalex.org/keywords/bias-mitigation","display_name":"bias mitigation","score":0.546999990940094},{"id":"https://openalex.org/keywords/conversational-partners","display_name":"conversation partners","score":0.414000004529953},{"id":"https://openalex.org/keywords/model-capacity","display_name":"model capacity","score":0.3619999885559082}],"concepts":[{"id":"https://openalex.org/C2777200299","wikidata":"https://www.wikidata.org/wiki/Q52943","display_name":"Conversation","level":2,"score":0.6868894100189209},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.6607540845870972},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.5945029854774475},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.5505764484405518},{"id":"https://openalex.org/C76509639","wikidata":"https://www.wikidata.org/wiki/Q918036","display_name":"Race (biology)","level":2,"score":0.5315525531768799},{"id":"https://openalex.org/C137403100","wikidata":"https://www.wikidata.org/wiki/Q41710","display_name":"Ethnic group","level":2,"score":0.522087037563324},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.49576863646507263},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4949330687522888},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.4852311611175537},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.4333834946155548},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.4203026294708252},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.40091681480407715},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.35593539476394653},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3348273038864136},{"id":"https://openalex.org/C46312422","wikidata":"https://www.wikidata.org/wiki/Q11024","display_name":"Communication","level":1,"score":0.17259269952774048},{"id":"https://openalex.org/C144024400","wikidata":"https://www.wikidata.org/wiki/Q21201","display_name":"Sociology","level":0,"score":0.16855841875076294},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","level":1,"score":0.0},{"id":"https://openalex.org/C107993555","wikidata":"https://www.wikidata.org/wiki/Q1662673","display_name":"Gender studies","level":1,"score":0.0},{"id":"https://openalex.org/C19165224","wikidata":"https://www.wikidata.org/wiki/Q23404","display_name":"Anthropology","level":1,"score":0.0},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2109.03300","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2109.03300","pdf_url":"https://arxiv.org/pdf/2109.03300","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.2109.03300","doi":"https://doi.org/10.48550/arxiv.2109.03300","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2109.03300","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":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2109.03300","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2109.03300","pdf_url":"https://arxiv.org/pdf/2109.03300","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":[{"score":0.4074000120162964,"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10"}],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":83,"referenced_works":["https://openalex.org/W64581525","https://openalex.org/W1528168582","https://openalex.org/W1538580550","https://openalex.org/W1604166575","https://openalex.org/W1970838780","https://openalex.org/W1999123028","https://openalex.org/W2008903655","https://openalex.org/W2018916157","https://openalex.org/W2038600245","https://openalex.org/W2065852373","https://openalex.org/W2081852155","https://openalex.org/W2119805155","https://openalex.org/W2130386000","https://openalex.org/W2333459074","https://openalex.org/W2468036577","https://openalex.org/W2483215953","https://openalex.org/W2753230086","https://openalex.org/W2761699783","https://openalex.org/W2791170418","https://openalex.org/W2798280648","https://openalex.org/W2799763234","https://openalex.org/W2885421725","https://openalex.org/W2891525068","https://openalex.org/W2893425640","https://openalex.org/W2903410490","https://openalex.org/W2911227954","https://openalex.org/W2913443447","https://openalex.org/W2922862135","https://openalex.org/W2926555354","https://openalex.org/W2947432670","https://openalex.org/W2947599143","https://openalex.org/W2951000191","https://openalex.org/W2962787423","https://openalex.org/W2962989446","https://openalex.org/W2963167310","https://openalex.org/W2963349562","https://openalex.org/W2963526187","https://openalex.org/W2963825865","https://openalex.org/W2970583189","https://openalex.org/W2971883198","https://openalex.org/W2972413484","https://openalex.org/W2972668795","https://openalex.org/W2978017171","https://openalex.org/W2981385260","https://openalex.org/W2983931018","https://openalex.org/W2987407680","https://openalex.org/W2988937804","https://openalex.org/W2995404354","https://openalex.org/W3016952972","https://openalex.org/W3019416653","https://openalex.org/W3020980524","https://openalex.org/W3022579796","https://openalex.org/W3023786569","https://openalex.org/W3032388710","https://openalex.org/W3034115845","https://openalex.org/W3034153429","https://openalex.org/W3035068109","https://openalex.org/W3037132330","https://openalex.org/W3037697022","https://openalex.org/W3088599783","https://openalex.org/W3089263616","https://openalex.org/W3089276844","https://openalex.org/W3089430725","https://openalex.org/W3091818438","https://openalex.org/W3093211917","https://openalex.org/W3093233911","https://openalex.org/W3094342783","https://openalex.org/W3100355408","https://openalex.org/W3105966348","https://openalex.org/W3117655171","https://openalex.org/W3123127232","https://openalex.org/W3123223998","https://openalex.org/W3128232076","https://openalex.org/W3133702157","https://openalex.org/W3148043792","https://openalex.org/W3154118109","https://openalex.org/W3154654049","https://openalex.org/W3156661774","https://openalex.org/W3156775540","https://openalex.org/W3170534228","https://openalex.org/W3172415559","https://openalex.org/W3173168772","https://openalex.org/W6912424950"],"related_works":["https://openalex.org/W2983931018","https://openalex.org/W3117655171","https://openalex.org/W3100355408","https://openalex.org/W4410890450","https://openalex.org/W4417509034","https://openalex.org/W4394780976","https://openalex.org/W7160302948","https://openalex.org/W7222953172","https://openalex.org/W7081974286","https://openalex.org/W4401833735"],"abstract_inverted_index":{"All":[0],"AI":[1],"models":[2,34,101],"are":[3,12,133],"susceptible":[4],"to":[5,33,84,103],"learning":[6],"biases":[7,68],"in":[8,48,135,138,151,154],"data":[9],"that":[10,35,98,117],"they":[11],"trained":[13,20],"on.":[14],"For":[15],"generative":[16],"dialogue":[17,78,123],"models,":[18,124],"being":[19],"on":[21,59,141],"real":[22],"human":[23],"conversations":[24,57,72,158],"containing":[25],"unbalanced":[26],"gender":[27,93,106],"and":[28,108,130],"race/ethnicity":[29],"references":[30],"can":[31],"lead":[32],"display":[36],"learned":[37],"biases,":[38],"which":[39],"we":[40],"define":[41],"here":[42],"broadly":[43],"as":[44],"any":[45],"measurable":[46],"differences":[47,153],"the":[49,64],"distributions":[50],"of":[51,56,66,76,111,120],"words":[52],"or":[53,167],"semantic":[54],"content":[55],"based":[58],"demographic":[60],"groups.":[61],"We":[62,96,115],"measure":[63],"strength":[65],"such":[67],"by":[69,113],"producing":[70],"artificial":[71],"between":[73],"two":[74],"copies":[75],"a":[77,86,91,142],"model,":[79],"conditioning":[80],"one":[81],"conversational":[82,144],"partner":[83],"state":[85],"name":[87,126],"commonly":[88],"associated":[89,163],"with":[90,164],"certain":[92],"and/or":[94],"race/ethnicity.":[95],"find":[97],"larger":[99],"capacity":[100],"tend":[102],"exhibit":[104],"more":[105],"bias":[107,137],"greater":[109],"stereotyping":[110],"occupations":[112],"gender.":[114],"show":[116],"several":[118],"methods":[119],"tuning":[121],"these":[122],"specifically":[125],"scrambling,":[127],"controlled":[128],"generation,":[129],"unlikelihood":[131],"training,":[132],"effective":[134,150],"reducing":[136],"conversation,":[139],"including":[140],"downstream":[143],"task.":[145],"Name":[146],"scrambling":[147],"is":[148],"also":[149],"lowering":[152],"token":[155],"usage":[156],"across":[157],"where":[159],"partners":[160],"have":[161],"names":[162],"different":[165],"genders":[166],"races/ethnicities.":[168]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":4},{"year":2022,"cited_by_count":6},{"year":2021,"cited_by_count":1}],"updated_date":"2026-10-16T09:59:22.235173","created_date":"2021-09-13T00:00:00"}