{"id":"https://openalex.org/W4416138652","doi":"https://doi.org/10.52202/085713-4292","title":"Quantifying Cross-Modality Memorization in Vision-Language Models","display_name":"Quantifying Cross-Modality Memorization in Vision-Language Models","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416138652","doi":"https://doi.org/10.52202/085713-4292"},"language":"en","primary_location":{"id":"doi:10.52202/085713-4292","doi":"https://doi.org/10.52202/085713-4292","is_oa":false,"landing_page_url":"https://doi.org/10.52202/085713-4292","pdf_url":null,"source":{"id":"https://openalex.org/S4363606243","display_name":"neural information processing systems","issn_l":"1049-5258","issn":["1049-5258"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","norway-2"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Advances in Neural Information Processing Systems 38","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","is_traditional":true,"indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2506.05198","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5058342436","display_name":"Yuxin Wen","orcid":"https://orcid.org/0000-0003-3719-9001","observed_orcids":["https://orcid.org/0000-0003-3719-9001"]},"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":"Yuxin Wen","raw_affiliation_strings":["University of Maryland"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Maryland","institution_ids":["https://openalex.org/I66946132"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082426637","display_name":"Yangsibo Huang","orcid":"https://orcid.org/0000-0002-0640-4845","observed_orcids":["https://orcid.org/0000-0002-0640-4845"]},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yangsibo Huang","raw_affiliation_strings":["Google"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060687985","display_name":"Tom Goldstein","orcid":"https://orcid.org/0000-0003-1660-9307","observed_orcids":["https://orcid.org/0000-0003-1660-9307"]},"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":"Tom Goldstein","raw_affiliation_strings":["University of Maryland"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Maryland","institution_ids":["https://openalex.org/I66946132"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101772779","display_name":"Ravi Kumar","orcid":"https://orcid.org/0000-0002-2203-2586","observed_orcids":["https://orcid.org/0000-0002-2203-2586"]},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ravi Kumar","raw_affiliation_strings":["Google"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102807325","display_name":"Badih Ghazi","orcid":"https://orcid.org/0000-0001-8254-3268","observed_orcids":["https://orcid.org/0000-0001-8254-3268","https://orcid.org/0009-0004-1555-5321"]},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Badih Ghazi","raw_affiliation_strings":["Google Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5016054994","display_name":"Chiyuan Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chiyuan Zhang","raw_affiliation_strings":["Google Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.1445,"has_fulltext":true,"cited_by_count":6,"citation_normalized_percentile":{"value":0.91295037,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":93,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"142986","last_page":"143008"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9932000041007996,"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":0.9932000041007996,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.003000000026077032,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.0005000000237487257,"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/vision-language-models","display_name":"vision-language models","score":0.9860000014305115},{"id":"https://openalex.org/keywords/machine-unlearning","display_name":"machine unlearning","score":0.7440000176429749},{"id":"https://openalex.org/keywords/multi-hop-reasoning","display_name":"multi-hop reasoning","score":0.6970000267028809},{"id":"https://openalex.org/keywords/multimodal-learning","display_name":"multimodal learning","score":0.5879999995231628},{"id":"https://openalex.org/keywords/knowledge-acquisition","display_name":"knowledge acquisition","score":0.33500000834465027}],"concepts":[{"id":"https://openalex.org/C30038468","wikidata":"https://www.wikidata.org/wiki/Q4354775","display_name":"Memorization","level":2,"score":0.9034000039100647},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6787999868392944},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6093000173568726},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.5580000281333923},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.5516999959945679},{"id":"https://openalex.org/C61272859","wikidata":"https://www.wikidata.org/wiki/Q7834031","display_name":"Transferability","level":3,"score":0.5002999901771545},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.4821000099182129},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4083999991416931},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.38029998540878296},{"id":"https://openalex.org/C2777601683","wikidata":"https://www.wikidata.org/wiki/Q6499736","display_name":"Vocabulary","level":2,"score":0.34929999709129333},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.34049999713897705},{"id":"https://openalex.org/C150899416","wikidata":"https://www.wikidata.org/wiki/Q1820378","display_name":"Transfer of learning","level":2,"score":0.3312000036239624},{"id":"https://openalex.org/C2778012447","wikidata":"https://www.wikidata.org/wiki/Q1034415","display_name":"Scope (computer science)","level":2,"score":0.31290000677108765},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.3098999857902527},{"id":"https://openalex.org/C158154518","wikidata":"https://www.wikidata.org/wiki/Q7310970","display_name":"Relevance (law)","level":2,"score":0.3073999881744385},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.26269999146461487},{"id":"https://openalex.org/C2780767217","wikidata":"https://www.wikidata.org/wiki/Q5532421","display_name":"Generality","level":2,"score":0.2615000009536743}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.52202/085713-4292","doi":"https://doi.org/10.52202/085713-4292","is_oa":false,"landing_page_url":"https://doi.org/10.52202/085713-4292","pdf_url":null,"source":{"id":"https://openalex.org/S4363606243","display_name":"neural information processing systems","issn_l":"1049-5258","issn":["1049-5258"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","norway-2"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Advances in Neural Information Processing Systems 38","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2506.05198","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2506.05198","pdf_url":"https://arxiv.org/pdf/2506.05198","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.2506.05198","doi":"https://doi.org/10.48550/arxiv.2506.05198","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2506.05198","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:2506.05198","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2506.05198","pdf_url":"https://arxiv.org/pdf/2506.05198","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":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416138652.pdf","grobid_xml":"https://content.openalex.org/works/W4416138652.grobid-xml"},"referenced_works_count":19,"referenced_works":["https://openalex.org/W2908510526","https://openalex.org/W2950700477","https://openalex.org/W3083410900","https://openalex.org/W3130178918","https://openalex.org/W4386555476","https://openalex.org/W4386977995","https://openalex.org/W4388926373","https://openalex.org/W4389216598","https://openalex.org/W4390041933","https://openalex.org/W4390833824","https://openalex.org/W4392822190","https://openalex.org/W4399912922","https://openalex.org/W4402427382","https://openalex.org/W4404396608","https://openalex.org/W4404782801","https://openalex.org/W4404821172","https://openalex.org/W4410357125","https://openalex.org/W4415796767","https://openalex.org/W6947991961"],"related_works":["https://openalex.org/W4404638652","https://openalex.org/W7172540899","https://openalex.org/W3200057130","https://openalex.org/W4404781081","https://openalex.org/W7165665271","https://openalex.org/W7169875549","https://openalex.org/W7213266381","https://openalex.org/W7134988712","https://openalex.org/W4404341809","https://openalex.org/W3197394194"],"abstract_inverted_index":{"Understanding":[0],"what":[1],"and":[2,22,80,105,113,123,154,173],"how":[3],"neural":[4],"networks":[5],"memorize":[6],"during":[7],"training":[8,117],"is":[9],"crucial,":[10],"both":[11],"from":[12,23],"the":[13,24,74,128,141,152,174,178],"perspective":[14],"of":[15,18,26,77],"unintentional":[16],"memorization":[17,39,47,54,79,112],"potentially":[19],"sensitive":[20],"information":[21,150],"standpoint":[25],"effective":[27],"knowledge":[28,111],"acquisition":[29],"for":[30],"real-world,":[31],"knowledge-intensive":[32],"tasks.":[33],"While":[34],"previous":[35],"studies":[36],"primarily":[37],"investigate":[38],"within":[40],"a":[41,82,96,120,144,182],"single":[42,121],"modality,":[43],"such":[44],"as":[45],"text":[46],"in":[48,55,65,127,136,151],"large":[49],"language":[50],"models":[51,60,118],"or":[52],"image":[53],"diffusion":[56],"models,":[57,170],"unified":[58],"multimodal":[59,201],"are":[61],"becoming":[62],"increasingly":[63],"prevalent":[64],"practical":[66],"applications.":[67],"In":[68],"this":[69,161,187],"work,":[70],"we":[71,93,158,180],"focus":[72],"on":[73,86,119,197],"unique":[75],"characteristics":[76],"cross-modality":[78],"conduct":[81],"systematic":[83],"study":[84,192],"centered":[85],"vision-language":[87],"models.":[88],"To":[89],"facilitate":[90],"controlled":[91],"experiments,":[92],"first":[94],"introduce":[95],"synthetic":[97,102],"persona":[98],"dataset":[99],"comprising":[100],"diverse":[101],"person":[103],"images":[104],"textual":[106],"descriptions.":[107],"We":[108,189],"quantify":[109],"factual":[110],"cross-modal":[114,206],"transferability":[115],"by":[116],"modality":[122,138],"evaluating":[124],"their":[125],"performance":[126],"other.":[129],"Our":[130],"results":[131],"reveal":[132],"that":[133,160],"facts":[134],"learned":[135],"one":[137],"transfer":[139],"to":[140,185,204],"other,":[142],"but":[143],"significant":[145],"gap":[146,162],"exists":[147,163],"between":[148],"recalling":[149],"source":[153],"target":[155],"modalities.":[156],"Furthermore,":[157],"observe":[159],"across":[164],"various":[165],"scenarios,":[166],"including":[167],"more":[168,199],"capable":[169],"machine":[171],"unlearning,":[172],"multi-hop":[175],"case.":[176],"At":[177],"end,":[179],"propose":[181],"baseline":[183],"method":[184],"mitigate":[186],"challenge.":[188],"hope":[190],"our":[191],"can":[193],"inspire":[194],"future":[195],"research":[196],"developing":[198],"robust":[200],"learning":[202],"techniques":[203],"enhance":[205],"transferability.":[207]},"counts_by_year":[{"year":2026,"cited_by_count":4},{"year":2025,"cited_by_count":2}],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2025-10-10T00:00:00"}