{"id":"https://openalex.org/W7128094516","doi":"https://doi.org/10.48550/arxiv.2602.05748","title":"LeakBoost: Perceptual-Loss-Based Membership Inference Attack","display_name":"LeakBoost: Perceptual-Loss-Based Membership Inference Attack","publication_year":2026,"publication_date":"2026-02-05","ids":{"openalex":"https://openalex.org/W7128094516","doi":"https://doi.org/10.48550/arxiv.2602.05748"},"language":null,"primary_location":{"id":"pmh:oai:arXiv.org:2602.05748","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2602.05748","pdf_url":"https://arxiv.org/pdf/2602.05748","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":"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"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2602.05748","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5125167731","display_name":"Amit Kravchik Taub","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Taub, Amit Kravchik","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055676548","display_name":"Fred M. Grabovski","orcid":"https://orcid.org/0009-0000-1652-7614","observed_orcids":["https://orcid.org/0009-0000-1652-7614"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Grabovski, Fred M.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5084959917","display_name":"Guy Amit","orcid":"https://orcid.org/0000-0002-5987-4402","observed_orcids":["https://orcid.org/0000-0002-5987-4402"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Amit, Guy","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5047995605","display_name":"Yisroel Mirsky","orcid":"https://orcid.org/0000-0001-6367-2734","observed_orcids":["https://orcid.org/0000-0001-6367-2734"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mirsky, Yisroel","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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.7562999725341797,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.7562999725341797,"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":0.2401999980211258,"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/T12357","display_name":"Digital Media Forensic Detection","score":0.0010000000474974513,"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/membership-inference-attacks","display_name":"membership inference attacks","score":0.8650000095367432},{"id":"https://openalex.org/keywords/privacy-leakage","display_name":"privacy leakage","score":0.4830000102519989},{"id":"https://openalex.org/keywords/perceptual-loss","display_name":"perceptual loss","score":0.7459999918937683},{"id":"https://openalex.org/keywords/activation-space","display_name":"activation space","score":0.4699999988079071},{"id":"https://openalex.org/keywords/neural-network-privacy","display_name":"neural network privacy","score":0.6439999938011169},{"id":"https://openalex.org/keywords/image-classification","display_name":"image classification","score":0.4169999957084656},{"id":"https://openalex.org/keywords/white-box-attacks","display_name":"white-box attacks","score":0.5659999847412109},{"id":"https://openalex.org/keywords/deep-neural-network","display_name":"deep neural network","score":0.35100001096725464}],"concepts":[{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.7360000014305115},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7207000255584717},{"id":"https://openalex.org/C2776240099","wikidata":"https://www.wikidata.org/wiki/Q327018","display_name":"Interrogation","level":2,"score":0.7121000289916992},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5918999910354614},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5662000179290771},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5493999719619751},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.545199990272522},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5029000043869019},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.4433000087738037},{"id":"https://openalex.org/C101468663","wikidata":"https://www.wikidata.org/wiki/Q1620158","display_name":"Modular design","level":2,"score":0.44209998846054077},{"id":"https://openalex.org/C21200559","wikidata":"https://www.wikidata.org/wiki/Q7451068","display_name":"Sensitivity (control systems)","level":2,"score":0.4203000068664551},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.399399995803833},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.373199999332428},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.3522000014781952},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.3246000111103058},{"id":"https://openalex.org/C25343380","wikidata":"https://www.wikidata.org/wiki/Q277521","display_name":"Relation (database)","level":2,"score":0.3156000077724457},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.2896000146865845},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.2800000011920929},{"id":"https://openalex.org/C2781121602","wikidata":"https://www.wikidata.org/wiki/Q3504403","display_name":"Modular neural network","level":4,"score":0.2709999978542328},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.26030001044273376},{"id":"https://openalex.org/C182306322","wikidata":"https://www.wikidata.org/wiki/Q1779371","display_name":"Order (exchange)","level":2,"score":0.25760000944137573}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2602.05748","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2602.05748","pdf_url":"https://arxiv.org/pdf/2602.05748","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":"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"},{"id":"doi:10.48550/arxiv.2602.05748","doi":"https://doi.org/10.48550/arxiv.2602.05748","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.05748","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2602.05748","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2602.05748","pdf_url":"https://arxiv.org/pdf/2602.05748","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":"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"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[{"score":0.7300726771354675,"id":"https://metadata.un.org/sdg/16","display_name":"Peace, justice, and strong institutions"}],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Membership":[0],"inference":[1,111],"attacks":[2],"(MIAs)":[3],"aim":[4],"to":[5,64,140,156,192],"determine":[6],"whether":[7],"a":[8,13,53,60,70,80,186],"sample":[9],"was":[10],"part":[11],"of":[12,45,153,202],"model's":[14,61],"training":[15],"set,":[16],"posing":[17],"serious":[18],"privacy":[19,194],"risks":[20,195],"for":[21],"modern":[22],"machine-learning":[23],"systems.":[24],"Existing":[25],"MIAs":[26],"primarily":[27],"rely":[28],"on":[29],"static":[30],"indicators,":[31],"such":[32],"as":[33],"loss":[34],"or":[35],"confidence,":[36],"and":[37,89,126,142,168,176,188],"do":[38],"not":[39],"fully":[40],"leverage":[41],"the":[42,103,173,200],"dynamic":[43,203],"behavior":[44],"models":[46],"when":[47],"actively":[48,58],"probed.":[49],"We":[50],"propose":[51],"LeakBoost,":[52],"perceptual-loss-based":[54],"interrogation":[55,76],"framework":[56],"that":[57,165,177],"probes":[59],"internal":[62],"representations":[63],"expose":[65],"hidden":[66],"membership":[67,99,110,204],"signals.":[68],"Given":[69],"candidate":[71],"input,":[72],"LeakBoost":[73,113,183],"synthesizes":[74],"an":[75,97,151],"image":[77,92,123],"by":[78,96,149],"optimizing":[79],"perceptual":[81],"(activation-space)":[82],"objective,":[83],"amplifying":[84],"representational":[85],"differences":[86],"between":[87],"members":[88],"non-members.":[90],"This":[91],"is":[93],"then":[94],"analyzed":[95],"off-the-shelf":[98],"detector,":[100],"without":[101],"modifying":[102],"detector":[104],"itself.":[105],"When":[106],"combined":[107],"with":[108],"existing":[109],"methods,":[112],"achieves":[114],"substantial":[115],"improvements":[116,178],"at":[117,145],"low":[118],"false-positive":[119],"rates":[120],"across":[121],"multiple":[122],"classification":[124],"datasets":[125],"diverse":[127],"neural":[128],"network":[129],"architectures.":[130],"In":[131],"particular,":[132],"it":[133],"raises":[134],"AUC":[135],"from":[136],"near-chance":[137],"levels":[138],"(0.53-0.62)":[139],"0.81-0.88,":[141],"increases":[143],"TPR":[144],"1":[146],"percent":[147],"FPR":[148],"over":[150],"order":[152],"magnitude":[154],"compared":[155],"strong":[157],"baseline":[158],"attacks.":[159],"A":[160],"detailed":[161],"sensitivity":[162],"analysis":[163],"reveals":[164],"deeper":[166],"layers":[167],"short,":[169],"low-learning-rate":[170],"optimization":[171],"produce":[172],"strongest":[174],"leakage,":[175],"concentrate":[179],"in":[180,196],"gradient-based":[181],"detectors.":[182],"thus":[184],"offers":[185],"modular":[187],"computationally":[189],"efficient":[190],"way":[191],"assess":[193],"white-box":[197],"settings,":[198],"advancing":[199],"study":[201],"inference.":[205]},"counts_by_year":[],"updated_date":"2026-09-29T14:09:16.559479","created_date":"2026-02-07T00:00:00"}