{"id":"https://openalex.org/W7161131318","doi":"https://doi.org/10.48550/arxiv.2605.12574","title":"DistractMIA: Black-Box Membership Inference on Vision-Language Models via Semantic Distraction","display_name":"DistractMIA: Black-Box Membership Inference on Vision-Language Models via Semantic Distraction","publication_year":2026,"publication_date":"2026-05-12","ids":{"openalex":"https://openalex.org/W7161131318","doi":"https://doi.org/10.48550/arxiv.2605.12574"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.12574","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12574","pdf_url":null,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.12574","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5067831144","display_name":"Hongyi Tang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tang, Hongyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136127306","display_name":"Zhihao Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Zhihao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136131895","display_name":"Yi Yang","orcid":"https://orcid.org/0000-0002-6197-1527"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Yi","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":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.6086000204086304,"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":0.6086000204086304,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.08479999750852585,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.05649999901652336,"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/inference","display_name":"Inference","score":0.6962000131607056},{"id":"https://openalex.org/keywords/intuition","display_name":"Intuition","score":0.4986000061035156},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.4799000024795532},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.4458000063896179},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.3837999999523163},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.3303999900817871},{"id":"https://openalex.org/keywords/distraction","display_name":"Distraction","score":0.31700000166893005},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.29350000619888306}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7397000193595886},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6962000131607056},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6575999855995178},{"id":"https://openalex.org/C132010649","wikidata":"https://www.wikidata.org/wiki/Q189222","display_name":"Intuition","level":2,"score":0.4986000061035156},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4916999936103821},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.4799000024795532},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.47130000591278076},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.4458000063896179},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.3837999999523163},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.3303999900817871},{"id":"https://openalex.org/C2776378700","wikidata":"https://www.wikidata.org/wiki/Q3030775","display_name":"Distraction","level":2,"score":0.31700000166893005},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.29350000619888306},{"id":"https://openalex.org/C130318100","wikidata":"https://www.wikidata.org/wiki/Q2268914","display_name":"Semantic similarity","level":2,"score":0.2888999879360199},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.2856999933719635},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2818000018596649},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.2815999984741211},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.2809999883174896},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.28029999136924744},{"id":"https://openalex.org/C2779439875","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Natural language understanding","level":3,"score":0.2624000012874603},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.2590999901294708},{"id":"https://openalex.org/C2779843651","wikidata":"https://www.wikidata.org/wiki/Q7390335","display_name":"SIGNAL (programming language)","level":2,"score":0.2531000077724457}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.12574","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12574","pdf_url":null,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.12574","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12574","pdf_url":null,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/16","score":0.4519922733306885,"display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Vision-language":[0],"models":[1],"(VLMs)":[2],"are":[3,126],"trained":[4],"on":[5,49,65,81,142,185],"large-scale":[6],"image-text":[7],"corpora":[8],"that":[9,112,174],"may":[10],"contain":[11],"private,":[12],"copyrighted,":[13],"or":[14,56,164],"otherwise":[15],"sensitive":[16],"data,":[17],"motivating":[18],"membership":[19,44,148],"inference":[20,45],"as":[21],"a":[22,95,143,186],"tool":[23],"for":[24,31],"training-data":[25],"auditing.":[26],"This":[27,105],"is":[28,107],"especially":[29],"challenging":[30],"deployed":[32],"VLMs,":[33],"where":[34],"auditors":[35],"typically":[36],"observe":[37],"only":[38],"generated":[39,102],"textual":[40,152],"responses.":[41],"Existing":[42],"VLM":[43],"attacks":[46],"either":[47],"rely":[48],"probability-level":[50],"signals":[51],"unavailable":[52],"in":[53],"such":[54],"settings,":[55],"use":[57],"mask-based":[58],"semantic":[59,82,97],"prediction":[60],"tasks":[61],"whose":[62],"effectiveness":[63],"depends":[64],"object-centric":[66,193],"visual":[67,87],"assumptions.":[68],"To":[69,133],"address":[70],"these":[71],"limitations,":[72],"we":[73],"propose":[74],"DistractMIA,":[75],"an":[76],"output-only":[77,179],"black-box":[78],"framework":[79],"based":[80],"distraction.":[83],"Rather":[84],"than":[85],"removing":[86],"evidence,":[88],"DistractMIA":[89,138,175],"preserves":[90],"the":[91,110,119,131],"original":[92,120],"image,":[93],"inserts":[94],"known":[96],"distractor,":[98],"and":[99,146,157,171,180],"measures":[100],"how":[101],"responses":[103],"change.":[104],"design":[106],"motivated":[108],"by":[109],"intuition":[111],"member":[113],"samples":[114,125],"remain":[115],"more":[116,127],"anchored":[117],"to":[118],"image":[121],"semantics,":[122],"while":[123],"non-member":[124],"easily":[128],"redirected":[129],"toward":[130],"distractor.":[132],"make":[134],"this":[135],"signal":[136],"reliable,":[137],"calibrates":[139],"distractor":[140,158],"configurations":[141],"reference":[144],"set":[145],"derives":[147],"scores":[149],"from":[150],"repeated":[151],"generations,":[153],"capturing":[154],"response":[155],"stability":[156],"uptake":[159],"without":[160],"accessing":[161],"logits,":[162],"probabilities,":[163],"hidden":[165],"states.":[166],"Experiments":[167],"across":[168],"multiple":[169],"VLMs":[170],"benchmarks":[172],"show":[173],"consistently":[176],"outperforms":[177],"both":[178],"stronger-access":[181],"baselines.":[182],"Its":[183],"performance":[184],"medical":[187],"benchmark":[188],"further":[189],"demonstrates":[190],"applicability":[191],"beyond":[192],"natural":[194],"images.":[195]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-15T00:00:00"}