{"id":"https://openalex.org/W4395687308","doi":"https://doi.org/10.1145/3630106.3659039","title":"Embracing Diversity: Interpretable Zero-shot Classification Beyond One Vector Per Class","display_name":"Embracing Diversity: Interpretable Zero-shot Classification Beyond One Vector Per Class","publication_year":2024,"publication_date":"2024-06-03","ids":{"openalex":"https://openalex.org/W4395687308","doi":"https://doi.org/10.1145/3630106.3659039"},"language":"en","primary_location":{"id":"doi:10.1145/3630106.3659039","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3630106.3659039","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3630106.3659039","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The 2024 ACM Conference on Fairness, Accountability, and Transparency","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3630106.3659039","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5077078550","display_name":"Mazda Moayeri","orcid":"https://orcid.org/0009-0006-2842-5352"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mazda Moayeri","raw_affiliation_strings":["University of Maryland, United States of America"],"raw_orcid":"https://orcid.org/0009-0006-2842-5352","affiliations":[{"raw_affiliation_string":"University of Maryland, United States of America","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089819604","display_name":"Michael Rabbat","orcid":"https://orcid.org/0000-0003-0536-7904"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Michael Rabbat","raw_affiliation_strings":["Meta AI, Canada"],"raw_orcid":"https://orcid.org/0000-0003-0536-7904","affiliations":[{"raw_affiliation_string":"Meta AI, Canada","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011104591","display_name":"Mark Ibrahim","orcid":"https://orcid.org/0000-0003-4972-9620"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mark Ibrahim","raw_affiliation_strings":["Meta AI, USA"],"raw_orcid":"https://orcid.org/0000-0003-4972-9620","affiliations":[{"raw_affiliation_string":"Meta AI, USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5075501154","display_name":"Diane Bouchacourt","orcid":"https://orcid.org/0009-0001-9078-8836"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Diane Bouchacourt","raw_affiliation_strings":["Meta AI, Canada"],"raw_orcid":"https://orcid.org/0009-0001-9078-8836","affiliations":[{"raw_affiliation_string":"Meta AI, Canada","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2302","last_page":"2321"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9998999834060669,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9998999834060669,"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":0.9987999796867371,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9976000189781189,"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/computer-science","display_name":"Computer science","score":0.6954495310783386},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6629117727279663},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.561747133731842},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.5364504456520081},{"id":"https://openalex.org/keywords/debugging","display_name":"Debugging","score":0.5134071111679077},{"id":"https://openalex.org/keywords/diversity","display_name":"Diversity (politics)","score":0.4191092252731323},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.35660505294799805},{"id":"https://openalex.org/keywords/data-mining","display_name":"Data mining","score":0.32779964804649353}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6954495310783386},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6629117727279663},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.561747133731842},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5364504456520081},{"id":"https://openalex.org/C168065819","wikidata":"https://www.wikidata.org/wiki/Q845566","display_name":"Debugging","level":2,"score":0.5134071111679077},{"id":"https://openalex.org/C2781316041","wikidata":"https://www.wikidata.org/wiki/Q1230584","display_name":"Diversity (politics)","level":2,"score":0.4191092252731323},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.35660505294799805},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.32779964804649353},{"id":"https://openalex.org/C19165224","wikidata":"https://www.wikidata.org/wiki/Q23404","display_name":"Anthropology","level":1,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C144024400","wikidata":"https://www.wikidata.org/wiki/Q21201","display_name":"Sociology","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3630106.3659039","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3630106.3659039","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3630106.3659039","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The 2024 ACM Conference on Fairness, Accountability, and Transparency","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2404.16717","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2404.16717","pdf_url":"https://arxiv.org/pdf/2404.16717","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"doi:10.1145/3630106.3659039","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3630106.3659039","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3630106.3659039","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The 2024 ACM Conference on Fairness, Accountability, and Transparency","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4395687308.pdf"},"referenced_works_count":13,"referenced_works":["https://openalex.org/W12634471","https://openalex.org/W1948251820","https://openalex.org/W1977295328","https://openalex.org/W2108598243","https://openalex.org/W2533598788","https://openalex.org/W3159481202","https://openalex.org/W3198377975","https://openalex.org/W4312310776","https://openalex.org/W4379474753","https://openalex.org/W4386065884","https://openalex.org/W4386076179","https://openalex.org/W4390871773","https://openalex.org/W4390872508"],"related_works":["https://openalex.org/W4321442002","https://openalex.org/W2015265939","https://openalex.org/W2284072287","https://openalex.org/W2611067230","https://openalex.org/W2480201319","https://openalex.org/W2387706296","https://openalex.org/W2155788121","https://openalex.org/W4235469518","https://openalex.org/W362492756","https://openalex.org/W2294325978"],"abstract_inverted_index":{"Vision-language":[0],"models":[1,24],"enable":[2],"open-world":[3],"classification":[4,92,130,234],"of":[5,46,68,135,187,219,232],"objects":[6,29,38],"without":[7,120,251],"the":[8,77,117,230,244],"need":[9],"for":[10,107,166,191,197,240],"any":[11],"retraining.":[12,121],"While":[13],"this":[14,85,224],"zero-shot":[15,91,118,129,233],"paradigm":[16],"marks":[17],"a":[18,44,54,58,69,72,89,96,101,110,132,184,203,217,236],"significant":[19],"advance,":[20],"even":[21],"today\u2019s":[22],"best":[23],"exhibit":[25],"skewed":[26],"performance":[27],"when":[28],"are":[30],"dissimilar":[31],"from":[32,49],"their":[33],"typical":[34],"depiction.":[35],"Real":[36],"world":[37],"such":[39],"as":[40,146],"pears":[41],"appear":[42],"in":[43,57,116,243],"variety":[45],"forms":[47],"\u2014":[48,60],"diced":[50],"to":[51,71,83,103,169,183,189,193],"whole,":[52],"on":[53,76],"table":[55],"or":[56],"bowl":[59],"yet":[61],"standard":[62,128],"VLM":[63],"classifiers":[64],"map":[65],"all":[66],"instances":[67],"class":[70,78,111,210,238],"single":[73,97,237],"vector":[74,239],"based":[75],"label.":[79],"We":[80,99,122,176,222],"argue":[81],"that":[82],"represent":[84],"rich":[86],"diversity":[87,108,152,242],"within":[88,109],"class,":[90],"should":[93],"move":[94],"beyond":[95,235],"vector.":[98],"propose":[100],"method":[102,125,159,180],"encode":[104],"and":[105,142,173,208,246],"account":[106,190],"using":[112],"inferred":[113],"attributes,":[114],"still":[115],"setting":[119],"find":[123,178],"our":[124,158,179,220],"consistently":[126],"outperforms":[127],"over":[131],"large":[133,185],"suite":[134],"datasets":[136,149],"encompassing":[137],"hierarchies,":[138],"diverse":[139],"object":[140],"states,":[141],"real-world":[143],"geographic":[144],"diversity,":[145],"well":[147],"finer-grained":[148],"where":[150],"intra-class":[151],"may":[153],"be":[154,214],"less":[155],"prevalent.":[156],"Importantly,":[157],"is":[160],"inherently":[161],"interpretable,":[162],"offering":[163],"faithful":[164],"explanations":[165],"each":[167],"inference":[168],"facilitate":[170],"model":[171],"debugging":[172],"enhance":[174],"transparency.":[175],"also":[177],"scales":[181],"efficiently":[182],"number":[186],"attributes":[188],"diversity\u2014leading":[192],"more":[194],"accurate":[195],"predictions":[196],"atypical":[198],"instances.":[199],"Finally,":[200],"we":[201],"characterize":[202],"principled":[204],"trade-off":[205],"between":[206],"overall":[207],"worst":[209],"accuracy,":[211],"which":[212],"can":[213],"tuned":[215],"via":[216],"hyperparameter":[218],"method.":[221],"hope":[223],"work":[225],"spurs":[226],"further":[227],"research":[228],"into":[229],"promise":[231],"capturing":[241],"world,":[245],"building":[247],"transparent":[248],"AI":[249],"systems":[250],"compromising":[252],"performance.":[253]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2024-04-27T00:00:00"}