{"id":"https://openalex.org/W2959360485","doi":"https://doi.org/10.48550/arxiv.1812.08769","title":"What are the biases in my word embedding?","display_name":"What are the biases in my word embedding?","publication_year":2018,"publication_date":"2018-12-20","ids":{"openalex":"https://openalex.org/W2959360485","doi":"https://doi.org/10.48550/arxiv.1812.08769","mag":"2959360485"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:1812.08769","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1812.08769","pdf_url":"https://arxiv.org/pdf/1812.08769","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/1812.08769","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5059342404","display_name":"Nathaniel Swinger","orcid":"https://orcid.org/0009-0008-6248-5652","observed_orcids":["https://orcid.org/0009-0008-6248-5652"]},"institutions":[{"id":"https://openalex.org/I4210143528","display_name":"Lexington City Schools","ror":"https://ror.org/043f52r95","country_code":"US","type":"education","lineage":["https://openalex.org/I4210143528"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Swinger, Nathaniel","raw_affiliation_strings":["Lexington High School","Lexington High School, Lexington, MA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Lexington High School","institution_ids":["https://openalex.org/I4210143528"]},{"raw_affiliation_string":"Lexington High School, Lexington, MA, USA","institution_ids":["https://openalex.org/I4210143528"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014101282","display_name":"Maria De\u2010Arteaga","orcid":"https://orcid.org/0000-0003-2297-3308","observed_orcids":["https://orcid.org/0000-0003-2297-3308"]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"De-Arteaga, Maria","raw_affiliation_strings":["Carnegie Mellon University","Carnegie Mellon University Pittsburgh PA USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]},{"raw_affiliation_string":"Carnegie Mellon University Pittsburgh PA USA","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078998782","display_name":"Neil Thomas Heffernan","orcid":"https://orcid.org/0000-0002-3280-288X","observed_orcids":["https://orcid.org/0000-0002-3280-288X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Heffernan, Neil Thomas","raw_affiliation_strings":["Shrewsbury High School, Shrewsbury, MA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shrewsbury High School, Shrewsbury, MA, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089144841","display_name":"Mark D.M. Leiserson","orcid":"https://orcid.org/0000-0002-1034-4363","observed_orcids":["https://orcid.org/0000-0002-1034-4363"]},"institutions":[{"id":"https://openalex.org/I66946132","display_name":"University of Maryland, College Park","ror":"https://ror.org/047s2c258","country_code":"US","type":"education","lineage":["https://openalex.org/I66946132"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Leiserson, Mark DM","raw_affiliation_strings":["University of Maryland","University of Maryland, Flibbertigibbet, MD, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Maryland","institution_ids":["https://openalex.org/I66946132"]},{"raw_affiliation_string":"University of Maryland, Flibbertigibbet, MD, USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5086904495","display_name":"Adam Tauman Kalai","orcid":"https://orcid.org/0000-0002-4559-8574","observed_orcids":["https://orcid.org/0000-0002-4559-8574"]},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kalai, Adam Tauman","raw_affiliation_strings":["Microsoft Research","[Microsoft Research, Cambridge, MA, USA]"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I1290206253"]},{"raw_affiliation_string":"[Microsoft Research, Cambridge, MA, USA]","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":23,"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":0.9384999871253967,"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":0.9384999871253967,"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/T12262","display_name":"Hate Speech and Cyberbullying Detection","score":0.027000000700354576,"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":0.008999999612569809,"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/word-embeddings","display_name":"word embeddings","score":0.9599999785423279},{"id":"https://openalex.org/keywords/algorithmic-bias","display_name":"algorithmic bias","score":0.8429999947547913},{"id":"https://openalex.org/keywords/gender-bias","display_name":"gender bias","score":0.5109999775886536},{"id":"https://openalex.org/keywords/racial-bias","display_name":"racial bias","score":0.9589999914169312},{"id":"https://openalex.org/keywords/intersectionality","display_name":"intersectionality","score":0.6159999966621399},{"id":"https://openalex.org/keywords/proxy-bias","display_name":"proxy bias","score":0.5329999923706055},{"id":"https://openalex.org/keywords/names","display_name":"names","score":0.42399999499320984},{"id":"https://openalex.org/keywords/sensitive-attributes","display_name":"sensitive attributes","score":0.5690000057220459}],"concepts":[{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.763455331325531},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7110685706138611},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.6345045566558838},{"id":"https://openalex.org/C62230096","wikidata":"https://www.wikidata.org/wiki/Q275969","display_name":"Crowdsourcing","level":2,"score":0.6190546154975891},{"id":"https://openalex.org/C2777462759","wikidata":"https://www.wikidata.org/wiki/Q18395344","display_name":"Word embedding","level":3,"score":0.5925226807594299},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5360392332077026},{"id":"https://openalex.org/C2780148112","wikidata":"https://www.wikidata.org/wiki/Q1432581","display_name":"Proxy (statistics)","level":2,"score":0.5108954310417175},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.4914904832839966},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4443328380584717},{"id":"https://openalex.org/C2992700788","wikidata":"https://www.wikidata.org/wiki/Q8461","display_name":"Racial bias","level":3,"score":0.41975852847099304},{"id":"https://openalex.org/C76509639","wikidata":"https://www.wikidata.org/wiki/Q918036","display_name":"Race (biology)","level":2,"score":0.411678284406662},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.27106451988220215},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1684596836566925},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0},{"id":"https://openalex.org/C136764020","wikidata":"https://www.wikidata.org/wiki/Q466","display_name":"World Wide Web","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C59822182","wikidata":"https://www.wikidata.org/wiki/Q441","display_name":"Botany","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:1812.08769","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1812.08769","pdf_url":"https://arxiv.org/pdf/1812.08769","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.1812.08769","doi":"https://doi.org/10.48550/arxiv.1812.08769","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.1812.08769","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:1812.08769","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1812.08769","pdf_url":"https://arxiv.org/pdf/1812.08769","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":[{"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10","score":0.7307999730110168}],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2959360485.pdf","grobid_xml":"https://content.openalex.org/works/W2959360485.grobid-xml"},"referenced_works_count":22,"referenced_works":["https://openalex.org/W1981830476","https://openalex.org/W1998043162","https://openalex.org/W2053860285","https://openalex.org/W2062109106","https://openalex.org/W2079320540","https://openalex.org/W2100960835","https://openalex.org/W2101234009","https://openalex.org/W2110065044","https://openalex.org/W2140534852","https://openalex.org/W2153579005","https://openalex.org/W2162670686","https://openalex.org/W2250539671","https://openalex.org/W2483215953","https://openalex.org/W2768894107","https://openalex.org/W2769358515","https://openalex.org/W2769580234","https://openalex.org/W2770618123","https://openalex.org/W2788481061","https://openalex.org/W2794635328","https://openalex.org/W2893425640","https://openalex.org/W2909858225","https://openalex.org/W2962772361"],"related_works":["https://openalex.org/W2998368506","https://openalex.org/W3016719704","https://openalex.org/W2913897682","https://openalex.org/W3013854004","https://openalex.org/W2952929029","https://openalex.org/W4281553050","https://openalex.org/W4402615953","https://openalex.org/W3165942956","https://openalex.org/W2952349219","https://openalex.org/W2483215953"],"abstract_inverted_index":{"This":[0,87],"paper":[1],"presents":[2],"an":[3],"algorithm":[4,12,72,140],"for":[5],"enumerating":[6],"biases":[7,39,125,166],"in":[8,42,58,168],"word":[9,49,59,151,181],"embeddings.":[10,50],"The":[11,51,71,136],"exposes":[13],"a":[14,34,142,150,155],"large":[15],"number":[16,156],"of":[17,44,48,94,133,144,157,175,185],"offensive":[18],"associations":[19,52],"related":[20],"to":[21,84,100,112,123,138],"sensitive":[22,82,134],"features":[23,83],"such":[24],"as":[25,96],"race":[26],"and":[27,67,117,149,183,187],"gender":[28],"on":[29,107,131,178],"publicly":[30,179],"available":[31,180],"embeddings,":[32],"including":[33],"supposedly":[35],"\"debiased\"":[36],"embedding.":[37,152],"These":[38],"are":[40,53,141],"concerning":[41],"light":[43],"the":[45,81,108,169,173],"widespread":[46],"use":[47],"identified":[54],"by":[55],"geometric":[56],"patterns":[57],"embeddings":[60,182],"that":[61,103,163],"run":[62],"parallel":[63],"between":[64],"people's":[65],"names":[66,198],"common":[68],"lower-case":[69],"tokens.":[70],"is":[73,88],"highly":[74],"unsupervised:":[75],"it":[76,119,121],"does":[77],"not":[78,200],"even":[79],"require":[80],"be":[85],"pre-specified.":[86],"desirable":[89],"because:":[90],"(a)":[91],"many":[92],"forms":[93],"discrimination--such":[95],"racial":[97],"discrimination--are":[98],"linked":[99],"social":[101],"constructs":[102],"may":[104,199],"vary":[105],"depending":[106],"context,":[109],"rather":[110],"than":[111],"categories":[113],"with":[114],"fixed":[115],"definitions;":[116],"(b)":[118],"makes":[120],"easier":[122],"identify":[124],"against":[126],"intersectional":[127],"groups,":[128],"which":[129],"depend":[130],"combinations":[132],"features.":[135],"inputs":[137],"our":[139,176],"list":[143],"target":[145],"tokens,":[146],"e.g.":[147],"names,":[148,186],"It":[153],"outputs":[154],"Word":[158],"Embedding":[159],"Association":[160],"Tests":[161],"(WEATs)":[162],"capture":[164],"various":[165],"present":[167],"data.":[170],"We":[171,193],"illustrate":[172],"utility":[174],"approach":[177],"lists":[184],"evaluate":[188],"its":[189],"output":[190],"using":[191],"crowdsourcing.":[192],"also":[194],"show":[195],"how":[196],"removing":[197],"remove":[201],"potential":[202],"proxy":[203],"bias.":[204]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":7},{"year":2022,"cited_by_count":4},{"year":2021,"cited_by_count":2},{"year":2020,"cited_by_count":6},{"year":2019,"cited_by_count":3}],"updated_date":"2026-11-03T10:50:19.383937","created_date":"2019-07-23T00:00:00"}