{"id":"https://openalex.org/W7127334578","doi":"https://doi.org/10.48550/arxiv.2602.01150","title":"SMI: Statistical Membership Inference for Reliable Unlearned Model Auditing","display_name":"SMI: Statistical Membership Inference for Reliable Unlearned Model Auditing","publication_year":2026,"publication_date":"2026-02-01","ids":{"openalex":"https://openalex.org/W7127334578","doi":"https://doi.org/10.48550/arxiv.2602.01150"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2602.01150","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Article"},"type":"article","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":null,"any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5124890398","display_name":"Jialong Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Jialong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027049671","display_name":"Zeming Wei","orcid":"https://orcid.org/0000-0002-9867-3929"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei, Zeming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5124885868","display_name":"Jiaxuan Zou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zou, Jiaxuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000589326","display_name":"Jiacheng Gong","orcid":"https://orcid.org/0009-0002-7986-2203"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gong, Jiacheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Fu, Jie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fu, Jie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5124947350","display_name":"Chengyang Dong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dong, Chengyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Xu, Heng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Heng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5124948004","display_name":"Jialong Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Jialong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"Liu, Bo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Bo","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":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.1239227,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"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":0.46700000762939453,"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":0.46700000762939453,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.1662999987602234,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.15000000596046448,"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/audit","display_name":"Audit","score":0.661300003528595},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.641700029373169},{"id":"https://openalex.org/keywords/reliability","display_name":"Reliability (semiconductor)","score":0.6363999843597412},{"id":"https://openalex.org/keywords/shadow","display_name":"Shadow (psychology)","score":0.5408999919891357},{"id":"https://openalex.org/keywords/forgetting","display_name":"Forgetting","score":0.5238999724388123},{"id":"https://openalex.org/keywords/statistical-inference","display_name":"Statistical inference","score":0.5117999911308289},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4763000011444092},{"id":"https://openalex.org/keywords/fiducial-inference","display_name":"Fiducial inference","score":0.43939998745918274}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6952000260353088},{"id":"https://openalex.org/C199521495","wikidata":"https://www.wikidata.org/wiki/Q181487","display_name":"Audit","level":2,"score":0.661300003528595},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.641700029373169},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.6363999843597412},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5623000264167786},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5449000000953674},{"id":"https://openalex.org/C117797892","wikidata":"https://www.wikidata.org/wiki/Q286363","display_name":"Shadow (psychology)","level":2,"score":0.5408999919891357},{"id":"https://openalex.org/C7149132","wikidata":"https://www.wikidata.org/wiki/Q1377840","display_name":"Forgetting","level":2,"score":0.5238999724388123},{"id":"https://openalex.org/C134261354","wikidata":"https://www.wikidata.org/wiki/Q938438","display_name":"Statistical inference","level":2,"score":0.5117999911308289},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4763000011444092},{"id":"https://openalex.org/C95167961","wikidata":"https://www.wikidata.org/wiki/Q4483495","display_name":"Fiducial inference","level":5,"score":0.43939998745918274},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.39640000462532043},{"id":"https://openalex.org/C87007009","wikidata":"https://www.wikidata.org/wiki/Q210832","display_name":"Statistical hypothesis testing","level":2,"score":0.39629998803138733},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3677000105381012},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.367000013589859},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.35589998960494995},{"id":"https://openalex.org/C197640229","wikidata":"https://www.wikidata.org/wiki/Q2534066","display_name":"Predictability","level":2,"score":0.30239999294281006},{"id":"https://openalex.org/C201729545","wikidata":"https://www.wikidata.org/wiki/Q2193887","display_name":"Reliability theory","level":3,"score":0.28200000524520874},{"id":"https://openalex.org/C48057537","wikidata":"https://www.wikidata.org/wiki/Q3551145","display_name":"Statistical theory","level":2,"score":0.2775000035762787},{"id":"https://openalex.org/C45235069","wikidata":"https://www.wikidata.org/wiki/Q278425","display_name":"Table (database)","level":2,"score":0.2720000147819519},{"id":"https://openalex.org/C205167067","wikidata":"https://www.wikidata.org/wiki/Q3300636","display_name":"Interval estimation","level":3,"score":0.26980000734329224},{"id":"https://openalex.org/C96608239","wikidata":"https://www.wikidata.org/wiki/Q1199823","display_name":"Statistical power","level":2,"score":0.25600001215934753}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2602.01150","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Article"},{"id":"doi:10.48550/arxiv.2602.01150","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.01150","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,"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:doi:10.48550/arxiv.2602.01150","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Article"},"sustainable_development_goals":[{"score":0.6838204860687256,"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Machine":[0],"unlearning":[1,102],"(MU)":[2],"is":[3,21,61],"essential":[4],"for":[5,42,108,153],"enforcing":[6],"the":[7,82,133,138,144],"right":[8],"to":[9,23,97,181],"be":[10],"forgotten":[11,31],"in":[12,81,137],"machine":[13],"learning":[14],"systems.":[15],"A":[16],"key":[17],"challenge":[18],"of":[19,101],"MU":[20],"how":[22],"reliably":[24],"audit":[25],"whether":[26],"a":[27,124,176],"model":[28,44,170],"has":[29],"truly":[30],"specified":[32],"training":[33,105,171],"data.":[34],"Membership":[35,121],"Inference":[36,122],"Attacks":[37],"(MIAs)":[38],"are":[39,52],"widely":[40],"used":[41],"unlearned":[43,75,139],"auditing,":[45],"where":[46],"samples":[47,76],"that":[48,74,128,160],"evade":[49],"membership":[50,65],"detection":[51],"regarded":[53],"as":[54,131],"successfully":[55],"forgotten.":[56],"We":[57,72],"show":[58,159],"this":[59,89],"assumption":[60],"fundamentally":[62,78],"flawed:":[63],"failed":[64],"inference":[66],"does":[67],"not":[68],"imply":[69],"true":[70],"forgetting.":[71],"prove":[73],"occupy":[77],"different":[79],"positions":[80],"feature":[83,140],"space":[84],"than":[85],"non-member":[86,134],"samples,":[87],"making":[88],"alignment":[90],"bias":[91],"unavoidable":[92],"and":[93,178,189],"unobservable,":[94],"which":[95],"leads":[96],"systematically":[98],"optimistic":[99],"evaluations":[100],"performance.":[103,192],"Meanwhile,":[104],"shadow":[106,169],"models":[107],"MIA":[109],"incurs":[110],"substantial":[111],"computational":[112],"overhead.":[113],"To":[114],"address":[115],"both":[116,186],"limitations,":[117],"we":[118],"propose":[119],"Statistical":[120],"(SMI),":[123],"training-free":[125],"auditing":[126,130,155,183],"framework":[127],"reformulates":[129],"estimating":[132,143],"mixture":[135],"proportion":[136],"distribution.":[141],"Beyond":[142],"forgetting":[145],"rate,":[146],"SMI":[147,161,174],"also":[148],"provides":[149],"bootstrap":[150],"reference":[151],"ranges":[152],"quantified":[154],"reliability.":[156],"Extensive":[157],"experiments":[158],"consistently":[162],"outperforms":[163],"all":[164],"MIA-based":[165,182],"baselines,":[166],"with":[167,185],"no":[168],"required.":[172],"Overall,":[173],"establishes":[175],"principled":[177],"efficient":[179],"alternative":[180],"methods,":[184],"theoretical":[187],"guarantees":[188],"strong":[190],"empirical":[191]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-02-04T00:00:00"}