{"id":"https://openalex.org/W7153255689","doi":"https://doi.org/10.48550/arxiv.2604.07386","title":"Label Leakage Attacks in Machine Unlearning: A Parameter and Inversion-Based Approach","display_name":"Label Leakage Attacks in Machine Unlearning: A Parameter and Inversion-Based Approach","publication_year":2026,"publication_date":"2026-04-08","ids":{"openalex":"https://openalex.org/W7153255689","doi":"https://doi.org/10.48550/arxiv.2604.07386"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.07386","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.07386","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":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.2604.07386","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5068146104","display_name":"Weidong Zheng","orcid":"https://orcid.org/0000-0002-9793-5763"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Weidong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075512625","display_name":"Kongyang Chen","orcid":"https://orcid.org/0000-0003-2439-3518"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Kongyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101407816","display_name":"Yao Huang","orcid":"https://orcid.org/0000-0001-6931-8963"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Yao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133374030","display_name":"Yuanwei Guo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guo, Yuanwei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5133332932","display_name":"Yatie Xiao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiao, Yatie","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":0.8823000192642212,"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":0.8823000192642212,"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":0.0210999995470047,"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":0.014700000174343586,"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/discriminative-model","display_name":"Discriminative model","score":0.8194000124931335},{"id":"https://openalex.org/keywords/decision-tree","display_name":"Decision tree","score":0.5078999996185303},{"id":"https://openalex.org/keywords/entropy","display_name":"Entropy (arrow of time)","score":0.4860000014305115},{"id":"https://openalex.org/keywords/conditional-entropy","display_name":"Conditional entropy","score":0.415800005197525},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.40220001339912415},{"id":"https://openalex.org/keywords/experimental-data","display_name":"Experimental data","score":0.3856000006198883},{"id":"https://openalex.org/keywords/identification","display_name":"Identification (biology)","score":0.38260000944137573},{"id":"https://openalex.org/keywords/leakage","display_name":"Leakage (economics)","score":0.34200000762939453},{"id":"https://openalex.org/keywords/information-leakage","display_name":"Information leakage","score":0.33480000495910645}],"concepts":[{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.8194000124931335},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.703499972820282},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5957000255584717},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5095000267028809},{"id":"https://openalex.org/C84525736","wikidata":"https://www.wikidata.org/wiki/Q831366","display_name":"Decision tree","level":2,"score":0.5078999996185303},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.5011000037193298},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.4860000014305115},{"id":"https://openalex.org/C101721835","wikidata":"https://www.wikidata.org/wiki/Q813908","display_name":"Conditional entropy","level":3,"score":0.415800005197525},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.40220001339912415},{"id":"https://openalex.org/C55037315","wikidata":"https://www.wikidata.org/wiki/Q5421151","display_name":"Experimental data","level":2,"score":0.3856000006198883},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.38260000944137573},{"id":"https://openalex.org/C2777042071","wikidata":"https://www.wikidata.org/wiki/Q6509304","display_name":"Leakage (economics)","level":2,"score":0.34200000762939453},{"id":"https://openalex.org/C2779201187","wikidata":"https://www.wikidata.org/wiki/Q2775060","display_name":"Information leakage","level":2,"score":0.33480000495910645},{"id":"https://openalex.org/C12267149","wikidata":"https://www.wikidata.org/wiki/Q282453","display_name":"Support vector machine","level":2,"score":0.32269999384880066},{"id":"https://openalex.org/C123201435","wikidata":"https://www.wikidata.org/wiki/Q456632","display_name":"Information privacy","level":2,"score":0.31679999828338623},{"id":"https://openalex.org/C75553542","wikidata":"https://www.wikidata.org/wiki/Q178161","display_name":"A priori and a posteriori","level":2,"score":0.3111000061035156},{"id":"https://openalex.org/C65856478","wikidata":"https://www.wikidata.org/wiki/Q3991682","display_name":"Attack model","level":2,"score":0.3005000054836273},{"id":"https://openalex.org/C1893757","wikidata":"https://www.wikidata.org/wiki/Q3653001","display_name":"Inversion (geology)","level":3,"score":0.29679998755455017},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.28189998865127563},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.27549999952316284},{"id":"https://openalex.org/C131426668","wikidata":"https://www.wikidata.org/wiki/Q849203","display_name":"Electronic countermeasure","level":3,"score":0.272599995136261},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.26649999618530273},{"id":"https://openalex.org/C9679016","wikidata":"https://www.wikidata.org/wiki/Q1417473","display_name":"Principle of maximum entropy","level":2,"score":0.26570001244544983},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.26179999113082886},{"id":"https://openalex.org/C137822555","wikidata":"https://www.wikidata.org/wiki/Q2587068","display_name":"Information sensitivity","level":2,"score":0.2590999901294708},{"id":"https://openalex.org/C184297639","wikidata":"https://www.wikidata.org/wiki/Q177765","display_name":"Biometrics","level":2,"score":0.25519999861717224},{"id":"https://openalex.org/C3017597292","wikidata":"https://www.wikidata.org/wiki/Q25052250","display_name":"Privacy protection","level":2,"score":0.2542000114917755}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.07386","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.07386","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":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.2604.07386","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.07386","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"display_name":"Reduced inequalities","score":0.6783573627471924,"id":"https://metadata.un.org/sdg/10"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"With":[0],"the":[1,23,46,55,60,78,94,112,115,152,156,198,203,220,236,240],"widespread":[2],"application":[3],"of":[4,48,63,66,80,96,114,120,126,151,180,219,224],"artificial":[5],"intelligence":[6],"technologies":[7,35],"in":[8,69],"face":[9],"recognition":[10],"and":[11,72,83,118,129,143,167,191,222],"other":[12],"fields,":[13],"data":[14,68,128],"privacy":[15,31],"security":[16],"issues":[17],"have":[18,36],"received":[19],"extensive":[20],"attention,":[21],"especially":[22],"\\textit{right":[24],"to":[25,147,173,196],"be":[26],"forgotten}":[27],"emphasized":[28],"by":[29,103,239],"numerous":[30],"protection":[32],"laws.":[33],"Existing":[34],"proposed":[37,204],"various":[38],"unlearning":[39,57,213],"methods,":[40],"but":[41],"they":[42],"may":[43],"inadvertently":[44],"leak":[45],"categories":[47],"unlearned":[49,67],"data.":[50],"This":[51],"paper":[52],"focuses":[53],"on":[54,87,124,206],"category":[56,64],"scenario,":[58],"analyzes":[59],"potential":[61],"problems":[62],"leakage":[65],"multiple":[70],"scenarios,":[71],"proposes":[73],"four":[74,207],"attack":[75,166,172],"methods":[76],"from":[77],"perspectives":[79],"model":[81,84,97,117,157],"parameters":[82,113],"inversion":[85,158],"based":[86],"attackers":[88],"with":[89],"different":[90],"knowledge":[91],"backgrounds.":[92],"At":[93],"level":[95],"parameters,":[98],"we":[99,160],"construct":[100],"discriminative":[101],"features":[102,134],"computing":[104],"either":[105],"dot":[106],"products":[107],"or":[108],"vector":[109],"differences":[110],"between":[111],"target":[116,241],"those":[119],"auxiliary":[121],"models":[122],"trained":[123],"subsets":[125],"retained":[127],"unrelated":[130],"data,":[131],"respectively.":[132],"These":[133],"are":[135,184],"then":[136],"processed":[137],"via":[138],"k-means":[139],"clustering,":[140],"Youden's":[141],"Index,":[142],"decision":[144],"tree":[145],"algorithms":[146],"achieve":[148],"accurate":[149],"identification":[150],"forgotten":[153,199,238],"class.":[154,200],"In":[155],"domain,":[159],"design":[161],"a":[162,168,188,216],"gradient":[163],"optimization-based":[164],"white-box":[165],"genetic":[169],"algorithm-based":[170],"black-box":[171],"reconstruct":[174],"class-prototypical":[175],"samples.":[176],"The":[177],"prediction":[178],"profiles":[179],"these":[181],"synthesized":[182],"samples":[183],"subsequently":[185],"analyzed":[186],"using":[187],"threshold":[189],"criterion":[190,195],"an":[192],"information":[193],"entropy":[194],"infer":[197,235],"We":[201],"evaluate":[202],"attacks":[205],"standard":[208],"datasets":[209],"against":[210],"five":[211],"state-of-the-art":[212],"algorithms,":[214],"providing":[215],"detailed":[217],"analysis":[218],"strengths":[221],"limitations":[223],"each":[225],"method.":[226],"Experimental":[227],"results":[228],"demonstrate":[229],"that":[230],"our":[231],"approach":[232],"can":[233],"effectively":[234],"classes":[237],"model.":[242]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-04-11T00:00:00"}