{"id":"https://openalex.org/W4389523674","doi":"https://doi.org/10.18653/v1/2023.findings-emnlp.270","title":"Debiasing Multimodal Models via Causal Information Minimization","display_name":"Debiasing Multimodal Models via Causal Information Minimization","publication_year":2023,"publication_date":"2023-01-01","ids":{"openalex":"https://openalex.org/W4389523674","doi":"https://doi.org/10.18653/v1/2023.findings-emnlp.270"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2023.findings-emnlp.270","doi":"https://doi.org/10.18653/v1/2023.findings-emnlp.270","is_oa":true,"landing_page_url":"http://dx.doi.org/10.18653/v1/2023.findings-emnlp.270","pdf_url":"https://aclanthology.org/2023.findings-emnlp.270.pdf","source":{"id":"https://openalex.org/S7407096085","display_name":"Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP)","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":"Findings of the Association for Computational Linguistics: EMNLP 2023","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","is_traditional":true,"indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://aclanthology.org/2023.findings-emnlp.270.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5063726130","display_name":"Vaidehi Patil","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I114027177","display_name":"University of North Carolina at Chapel Hill","ror":"https://ror.org/0130frc33","country_code":"US","type":"education","lineage":["https://openalex.org/I114027177"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Vaidehi Patil","raw_affiliation_strings":["UNC Chapel Hill"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UNC Chapel Hill","institution_ids":["https://openalex.org/I114027177"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048025332","display_name":"Adyasha Maharana","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I114027177","display_name":"University of North Carolina at Chapel Hill","ror":"https://ror.org/0130frc33","country_code":"US","type":"education","lineage":["https://openalex.org/I114027177"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Adyasha Maharana","raw_affiliation_strings":["UNC Chapel Hill"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UNC Chapel Hill","institution_ids":["https://openalex.org/I114027177"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5001987532","display_name":"Mohit Bansal","orcid":"https://orcid.org/0000-0001-5522-1351","observed_orcids":["https://orcid.org/0000-0001-5522-1351"]},"institutions":[{"id":"https://openalex.org/I114027177","display_name":"University of North Carolina at Chapel Hill","ror":"https://ror.org/0130frc33","country_code":"US","type":"education","lineage":["https://openalex.org/I114027177"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mohit Bansal","raw_affiliation_strings":["UNC Chapel Hill"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UNC Chapel Hill","institution_ids":["https://openalex.org/I114027177"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I114027177"],"apc_list":null,"apc_paid":null,"fwci":0.1133,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.37425532,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":88,"max":93},"biblio":{"volume":null,"issue":null,"first_page":"4108","last_page":"4123"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.29760000109672546,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.29760000109672546,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.227400004863739,"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/T10883","display_name":"Ethics and Social Impacts of AI","score":0.19789999723434448,"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":0.8289999961853027},{"id":"https://openalex.org/keywords/multimodal-models","display_name":"multimodal models","score":0.9869999885559082},{"id":"https://openalex.org/keywords/dataset-bias","display_name":"data bias","score":0.7509999871253967},{"id":"https://openalex.org/keywords/out-of-distribution-generalization","display_name":"out-of-distribution generalization","score":0.9179999828338623},{"id":"https://openalex.org/keywords/spurious-features","display_name":"spurious features","score":0.8880000114440918},{"id":"https://openalex.org/keywords/causal-diagrams","display_name":"causal graphs","score":0.5590000152587891}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7593468427658081},{"id":"https://openalex.org/C2779458634","wikidata":"https://www.wikidata.org/wiki/Q24963715","display_name":"Debiasing","level":2,"score":0.74908846616745},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.709231972694397},{"id":"https://openalex.org/C97256817","wikidata":"https://www.wikidata.org/wiki/Q1462316","display_name":"Spurious relationship","level":2,"score":0.6476108431816101},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.6253795623779297},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5961693525314331},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5359538197517395},{"id":"https://openalex.org/C158600405","wikidata":"https://www.wikidata.org/wiki/Q5054566","display_name":"Causal inference","level":2,"score":0.5113314986228943},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.4534012973308563},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.40498030185699463},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.14617693424224854},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.12807312607765198},{"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/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.0},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/v1/2023.findings-emnlp.270","doi":"https://doi.org/10.18653/v1/2023.findings-emnlp.270","is_oa":true,"landing_page_url":"http://dx.doi.org/10.18653/v1/2023.findings-emnlp.270","pdf_url":"https://aclanthology.org/2023.findings-emnlp.270.pdf","source":{"id":"https://openalex.org/S7407096085","display_name":"Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP)","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":"Findings of the Association for Computational Linguistics: EMNLP 2023","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2311.16941","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2311.16941","pdf_url":"https://arxiv.org/pdf/2311.16941","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.2311.16941","doi":"https://doi.org/10.48550/arxiv.2311.16941","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2311.16941","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":"doi:10.18653/v1/2023.findings-emnlp.270","doi":"https://doi.org/10.18653/v1/2023.findings-emnlp.270","is_oa":true,"landing_page_url":"http://dx.doi.org/10.18653/v1/2023.findings-emnlp.270","pdf_url":"https://aclanthology.org/2023.findings-emnlp.270.pdf","source":{"id":"https://openalex.org/S7407096085","display_name":"Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP)","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":"Findings of the Association for Computational Linguistics: EMNLP 2023","raw_type":"proceedings-article"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G2921176242","display_name":null,"funder_award_id":"W911NF2110220","funder_id":"https://openalex.org/F4320338281","funder_display_name":"Army Research Office"},{"id":"https://openalex.org/G5599941371","display_name":null,"funder_award_id":"N00014-23-1-2356","funder_id":"https://openalex.org/F4320337345","funder_display_name":"Office of Naval Research"},{"id":"https://openalex.org/G7357862116","display_name":null,"funder_award_id":"N66001-19-2-4031","funder_id":"https://openalex.org/F4320332180","funder_display_name":"Defense Advanced Research Projects Agency"},{"id":"https://openalex.org/G8215844777","display_name":null,"funder_award_id":"N66001-19","funder_id":"https://openalex.org/F4320332180","funder_display_name":"Defense Advanced Research Projects Agency"},{"id":"https://openalex.org/G8876996369","display_name":null,"funder_award_id":"N00014","funder_id":"https://openalex.org/F4320337345","funder_display_name":"Office of Naval Research"}],"funders":[{"id":"https://openalex.org/F4320332180","display_name":"Defense Advanced Research Projects Agency","ror":"https://ror.org/02caytj08"},{"id":"https://openalex.org/F4320337345","display_name":"Office of Naval Research","ror":"https://ror.org/00rk2pe57"},{"id":"https://openalex.org/F4320338281","display_name":"Army Research Office","ror":"https://ror.org/05epdh915"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4389523674.pdf","grobid_xml":"https://content.openalex.org/works/W4389523674.grobid-xml"},"referenced_works_count":54,"referenced_works":["https://openalex.org/W1200752817","https://openalex.org/W1501375624","https://openalex.org/W1933349210","https://openalex.org/W2041404167","https://openalex.org/W2333897677","https://openalex.org/W2463267937","https://openalex.org/W2471094925","https://openalex.org/W2486285194","https://openalex.org/W2542071751","https://openalex.org/W2560730294","https://openalex.org/W2593634001","https://openalex.org/W2771951981","https://openalex.org/W2796084947","https://openalex.org/W2890952061","https://openalex.org/W2899335602","https://openalex.org/W2934463783","https://openalex.org/W2950104027","https://openalex.org/W2955124656","https://openalex.org/W2963890019","https://openalex.org/W2964013391","https://openalex.org/W2970019270","https://openalex.org/W2970231061","https://openalex.org/W2970548373","https://openalex.org/W2996388127","https://openalex.org/W3016970897","https://openalex.org/W3034564653","https://openalex.org/W3035017890","https://openalex.org/W3035517717","https://openalex.org/W3102995547","https://openalex.org/W3104788521","https://openalex.org/W3109835285","https://openalex.org/W3113067643","https://openalex.org/W3120172934","https://openalex.org/W3121939794","https://openalex.org/W3177934633","https://openalex.org/W3206381016","https://openalex.org/W3207736102","https://openalex.org/W3213504576","https://openalex.org/W3214559542","https://openalex.org/W4213057961","https://openalex.org/W4214825151","https://openalex.org/W4221146508","https://openalex.org/W4224920305","https://openalex.org/W4226392681","https://openalex.org/W4283806310","https://openalex.org/W4302305855","https://openalex.org/W4312457137","https://openalex.org/W4312884841","https://openalex.org/W4313057368","https://openalex.org/W4385574060","https://openalex.org/W4386043921","https://openalex.org/W4386065611","https://openalex.org/W7133208411","https://openalex.org/W7133222829"],"related_works":["https://openalex.org/W4405717864","https://openalex.org/W7166852798","https://openalex.org/W4415059472","https://openalex.org/W4399039389","https://openalex.org/W4416035043","https://openalex.org/W7131070659","https://openalex.org/W4387968274","https://openalex.org/W4403791296","https://openalex.org/W4404025009","https://openalex.org/W6910720448"],"abstract_inverted_index":{"Most":[0],"existing":[1],"debiasing":[2,143],"methods":[3,120,144],"for":[4,32,55],"multimodal":[5,33,56,151],"models,":[6],"including":[7],"causal":[8,53,123],"intervention":[9],"and":[10,58,116,140],"inference":[11],"methods,":[12],"utilize":[13],"approximate":[14],"heuristics":[15],"to":[16,68,83,125,162],"represent":[17],"the":[18,70,87,100,106,132,141,164,175],"biases,":[19],"such":[20],"as":[21,113],"shallow":[22],"features":[23,31,74,91,103,112,168],"from":[24,49,93,128],"early":[25],"stages":[26],"of":[27,90,166,177],"training":[28],"or":[29],"unimodal":[30],"tasks":[34],"like":[35],"VQA,":[36],"etc.,":[37],"which":[38],"may":[39],"not":[40],"be":[41],"accurate.In":[42],"this":[43],"paper,":[44],"we":[45,157],"study":[46],"bias":[47,127],"arising":[48],"confounders":[50],"in":[51,169],"a":[52,60,80,94,159],"graph":[54],"data,":[57],"examine":[59],"novel":[61,160],"approach":[62],"that":[63,78,104,131,172],"leverages":[64],"causallymotivated":[65],"information":[66,77,88],"minimization":[67],"learn":[69,99],"confounder":[71,114,134],"representations.Robust":[72],"predictive":[73,102],"contain":[75],"diverse":[76],"helps":[79,98],"model":[81,97],"generalize":[82],"out-of-distribution":[84,146],"data.Hence,":[85],"minimizing":[86],"content":[89],"obtained":[92],"pretrained":[95],"biased":[96],"simplest":[101],"capture":[105,137],"underlying":[107],"data":[108],"distribution.We":[109],"treat":[110],"these":[111],"representations":[115,135],"use":[117],"them":[118],"via":[119],"motivated":[121],"by":[122],"theory":[124],"remove":[126],"models.We":[129],"find":[130],"learned":[133],"indeed":[136],"dataset":[138],"biases":[139],"proposed":[142,179],"improve":[145],"(OOD)":[147],"performance":[148],"on":[149],"multiple":[150],"datasets":[152],"without":[153],"sacrificing":[154],"in-distribution":[155],"performance.Additionally,":[156],"introduce":[158],"metric":[161],"quantify":[163],"sufficiency":[165],"spurious":[167],"models'":[170],"predictions":[171],"further":[173],"demonstrates":[174],"effectiveness":[176],"our":[178],"methods.":[180],"1":[181]},"counts_by_year":[{"year":2024,"cited_by_count":1}],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2023-12-11T00:00:00"}