{"id":"https://openalex.org/W4412944526","doi":"https://doi.org/10.18653/v1/2025.findings-acl.1372","title":"Can Vision Language Models Understand Mimed Actions?","display_name":"Can Vision Language Models Understand Mimed Actions?","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412944526","doi":"https://doi.org/10.18653/v1/2025.findings-acl.1372"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.1372","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1372","pdf_url":"https://aclanthology.org/2025.findings-acl.1372.pdf","source":null,"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: ACL 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.1372.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5032451595","display_name":"Hyundong Cho","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hyundong Justin Cho","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5080644357","display_name":"Spencer Lin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Spencer Lin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082028315","display_name":"Tejas Srinivasan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tejas Srinivasan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034254782","display_name":"Michael Saxon","orcid":"https://orcid.org/0000-0001-7306-5030"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Michael Saxon","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111134271","display_name":"Deuksin Kwon","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Deuksin Kwon","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Natali T. Chavez","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Natali T. Chavez","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5000874697","display_name":"Jonathan May","orcid":"https://orcid.org/0000-0002-5284-477X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jonathan May","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"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":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"26744","last_page":"26759"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11148","display_name":"Language, Metaphor, and Cognition","score":0.9085999727249146,"subfield":{"id":"https://openalex.org/subfields/3205","display_name":"Experimental and Cognitive Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T11148","display_name":"Language, Metaphor, and Cognition","score":0.9085999727249146,"subfield":{"id":"https://openalex.org/subfields/3205","display_name":"Experimental and Cognitive Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7061275243759155},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.38234931230545044},{"id":"https://openalex.org/keywords/human\u2013computer-interaction","display_name":"Human\u2013computer interaction","score":0.364189088344574},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.3485035300254822}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7061275243759155},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38234931230545044},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.364189088344574},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3485035300254822}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.1372","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1372","pdf_url":"https://aclanthology.org/2025.findings-acl.1372.pdf","source":null,"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: ACL 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.1372","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1372","pdf_url":"https://aclanthology.org/2025.findings-acl.1372.pdf","source":null,"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: ACL 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412944526.pdf","grobid_xml":"https://content.openalex.org/works/W4412944526.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W3204019825"],"abstract_inverted_index":{"Nonverbal":[0],"communication":[1],"(NVC)":[2],"plays":[3],"an":[4],"integral":[5],"role":[6],"in":[7,13,26],"human":[8,58,155],"language,":[9],"but":[10],"studying":[11],"NVC":[12,47],"general":[14],"is":[15,69],"challenging":[16],"because":[17],"of":[18,35,46,50,66,77,84,100,110,112,154],"its":[19],"broad":[20],"scope":[21],"and":[22,30,42,52,79,122,132],"also":[23],"high":[24],"variance":[25],"interpretation":[27,59],"among":[28],"individuals":[29],"cultures.However,":[31],"mime-the":[32],"theatrical":[33],"technique":[34],"suggesting":[36],"intent":[37],"using":[38],"only":[39],"gesture,":[40],"expression,":[41],"movement-is":[43],"a":[44,63,70,93],"subset":[45],"that":[48,62,129],"consists":[49,109],"explicit":[51],"embodied":[53],"actions":[54,68],"with":[55,104,115],"much":[56],"lower":[57],"variance.We":[60],"argue":[61],"solid":[64],"understanding":[65,153],"mimed":[67,102],"crucial":[71],"prerequisite":[72],"for":[73,124,146,149],"vision-language":[74],"models":[75,135],"capable":[76],"interpreting":[78],"commanding":[80],"more":[81,151],"subtle":[82],"aspects":[83],"NVC.Hence,":[85],"we":[86],"propose":[87],"Mime":[88],"Identification":[89],"Multimodal":[90],"Evaluation":[91],"(MIME),":[92],"novel":[94],"videobased":[95],"question":[96],"answering":[97],"benchmark":[98],"comprising":[99],"86":[101],"actions.Constructed":[103],"motion":[105],"capture":[106],"data,":[107],"MIME":[108],"variations":[111],"each":[113],"action":[114],"perturbations":[116],"applied":[117],"to":[118],"the":[119,144],"character,":[120],"background,":[121],"viewpoint":[123],"evaluating":[125],"recognition":[126],"robustness.We":[127],"find":[128],"both":[130],"open-weight":[131],"API-based":[133],"visionlanguage":[134],"perform":[136],"significantly":[137],"worse":[138],"than":[139],"humans":[140],"on":[141],"MIME,":[142],"motivating":[143],"need":[145],"increased":[147],"research":[148],"instilling":[150],"robust":[152],"gestures.":[156]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}