{"id":"https://openalex.org/W7160238706","doi":"https://doi.org/10.1109/wacv61042.2026.00033","title":"A Woman with a Knife or A Knife with a Woman? Measuring Directional Bias Amplification in Image Captions","display_name":"A Woman with a Knife or A Knife with a Woman? Measuring Directional Bias Amplification in Image Captions","publication_year":2026,"publication_date":"2026-03-06","ids":{"openalex":"https://openalex.org/W7160238706","doi":"https://doi.org/10.1109/wacv61042.2026.00033"},"language":null,"primary_location":{"id":"doi:10.1109/wacv61042.2026.00033","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv61042.2026.00033","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5103254573","display_name":"Rahul Nair","orcid":"https://orcid.org/0000-0002-1875-2409"},"institutions":[{"id":"https://openalex.org/I55732556","display_name":"Arizona State University","ror":"https://ror.org/03efmqc40","country_code":"US","type":"education","lineage":["https://openalex.org/I55732556"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rahul Nair","raw_affiliation_strings":["Arizona State University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Arizona State University","institution_ids":["https://openalex.org/I55732556"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028142661","display_name":"Bhanu Tokas","orcid":null},"institutions":[{"id":"https://openalex.org/I55732556","display_name":"Arizona State University","ror":"https://ror.org/03efmqc40","country_code":"US","type":"education","lineage":["https://openalex.org/I55732556"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bhanu Tokas","raw_affiliation_strings":["Arizona State University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Arizona State University","institution_ids":["https://openalex.org/I55732556"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5135372203","display_name":"Hannah Kerner","orcid":null},"institutions":[{"id":"https://openalex.org/I55732556","display_name":"Arizona State University","ror":"https://ror.org/03efmqc40","country_code":"US","type":"education","lineage":["https://openalex.org/I55732556"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hannah Kerner","raw_affiliation_strings":["Arizona State University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Arizona State University","institution_ids":["https://openalex.org/I55732556"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I55732556"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.50548543,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"255","last_page":"264"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.8097000122070312,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.8097000122070312,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T13310","display_name":"Subtitles and Audiovisual Media","score":0.05400000140070915,"subfield":{"id":"https://openalex.org/subfields/1203","display_name":"Language and Linguistics"},"field":{"id":"https://openalex.org/fields/12","display_name":"Arts and Humanities"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T11439","display_name":"Video Analysis and Summarization","score":0.010900000110268593,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/noise","display_name":"Noise (video)","score":0.3319000005722046},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.3273000121116638},{"id":"https://openalex.org/keywords/image-processing","display_name":"Image processing","score":0.3190000057220459},{"id":"https://openalex.org/keywords/closed-captioning","display_name":"Closed captioning","score":0.2685999870300293},{"id":"https://openalex.org/keywords/pixel","display_name":"Pixel","score":0.24369999766349792}],"concepts":[{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.6620000004768372},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6220999956130981},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5960999727249146},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.3319000005722046},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.3273000121116638},{"id":"https://openalex.org/C9417928","wikidata":"https://www.wikidata.org/wiki/Q1070689","display_name":"Image processing","level":3,"score":0.3190000057220459},{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.2685999870300293},{"id":"https://openalex.org/C160633673","wikidata":"https://www.wikidata.org/wiki/Q355198","display_name":"Pixel","level":2,"score":0.24369999766349792},{"id":"https://openalex.org/C119657128","wikidata":"https://www.wikidata.org/wiki/Q11633","display_name":"Photography","level":2,"score":0.24199999868869781},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.23810000717639923}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/wacv61042.2026.00033","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv61042.2026.00033","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Gender equality","score":0.7661665081977844,"id":"https://metadata.un.org/sdg/5"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":26,"referenced_works":["https://openalex.org/W1895577753","https://openalex.org/W2064675550","https://openalex.org/W2133459682","https://openalex.org/W2250539671","https://openalex.org/W2493916176","https://openalex.org/W2745461083","https://openalex.org/W2962787423","https://openalex.org/W2962858109","https://openalex.org/W2963078909","https://openalex.org/W2963084599","https://openalex.org/W2963349562","https://openalex.org/W3032340379","https://openalex.org/W3091588028","https://openalex.org/W3173683732","https://openalex.org/W4285491539","https://openalex.org/W4312300614","https://openalex.org/W4312537386","https://openalex.org/W4389519103","https://openalex.org/W4390873716","https://openalex.org/W4393146540","https://openalex.org/W4399528683","https://openalex.org/W4401043876","https://openalex.org/W4402660140","https://openalex.org/W4402727764","https://openalex.org/W4402917081","https://openalex.org/W7133193597"],"related_works":[],"abstract_inverted_index":{"When":[0],"we":[1],"train":[2],"models":[3,112],"on":[4,150],"biased":[5],"datasets,":[6,50],"they":[7,52],"not":[8],"only":[9,39,165],"reproduce":[10],"data":[11],"biases,":[12],"but":[13],"can":[14,108],"worsen":[15],"them":[16],"at":[17],"test":[18],"time":[19],"\u2014":[20],"a":[21,59,68,80,102,139],"phenomenon":[22],"called":[23],"bias":[24,30,37,70,89,144,170],"amplification.":[25],"Many":[26],"of":[27,58,88,143],"the":[28,55,86,156,164],"current":[29],"amplification":[31,38,71,90,145,171],"metrics":[32,44],"(e.g.,":[33],"BA\u2192,":[34],"DPA)":[35],"measure":[36,169],"in":[40,65,91,99,132,146,155,172],"classification":[41],"datasets.":[42],"These":[43],"are":[45],"ineffective":[46],"for":[47],"image":[48],"captioning":[49,92,111],"as":[51],"cannot":[53,84],"capture":[54],"language":[56],"semantics":[57],"caption.":[60],"Recent":[61],"work":[62],"introduced":[63],"Leakage":[64],"Captioning":[66,100],"(LIC),":[67],"language-aware":[69,103,133],"metric":[72,106,167],"that":[73,107,161],"understands":[74],"caption":[75],"semantics.":[76],"However,":[77],"LIC":[78],"has":[79,116],"crucial":[81],"limitation:":[82],"it":[83,123,137],"identify":[85,109],"source":[87],"models.":[93],"We":[94],"propose":[95],"Directional":[96],"Bias":[97],"Amplification":[98],"(DBAC),":[101],"and":[104,135,152],"directional":[105],"when":[110],"amplify":[113],"biases.":[114],"DBAC":[115,162],"two":[117],"more":[118,140],"improvements":[119],"over":[120],"LIC:":[121],"(1)":[122],"is":[124,163],"less":[125],"sensitive":[126],"to":[127,168],"sentence":[128],"encoders":[129],"(a":[130],"hyperparameter":[131],"metrics),":[134],"(2)":[136],"provides":[138],"accurate":[141],"estimate":[142],"captions.":[147,173],"Our":[148],"experiments":[149],"gender":[151],"race":[153],"attributes":[154],"COCO":[157],"captions":[158],"dataset":[159],"show":[160],"reliable":[166]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-05-06T00:00:00"}