{"id":"https://openalex.org/W7138853863","doi":"https://doi.org/10.48550/arxiv.2603.17623","title":"ARES: Scalable and Practical Gradient Inversion Attack in Federated Learning through Activation Recovery","display_name":"ARES: Scalable and Practical Gradient Inversion Attack in Federated Learning through Activation Recovery","publication_year":2026,"publication_date":"2026-03-18","ids":{"openalex":"https://openalex.org/W7138853863","doi":"https://doi.org/10.48550/arxiv.2603.17623"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.17623","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17623","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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.17623","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5089152129","display_name":"Zirui Gong","orcid":"https://orcid.org/0009-0008-6284-2234"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Gong, Zirui","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129789877","display_name":"Leo Yu Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Leo Yu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130190106","display_name":"Yanjun Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Yanjun","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130171688","display_name":"Viet Vo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vo, Viet","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130196828","display_name":"Tianqing Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Tianqing","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129881361","display_name":"Shirui Pan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pan, Shirui","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130140523","display_name":"Cong Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Cong","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5089152129"],"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.8748999834060669,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.8748999834060669,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.09309999644756317,"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/T10237","display_name":"Cryptography and Data Security","score":0.0035000001080334187,"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/scalability","display_name":"Scalability","score":0.7889000177383423},{"id":"https://openalex.org/keywords/federated-learning","display_name":"Federated learning","score":0.5627999901771545},{"id":"https://openalex.org/keywords/inversion","display_name":"Inversion (geology)","score":0.5080000162124634},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.3727000057697296},{"id":"https://openalex.org/keywords/synthetic-data","display_name":"Synthetic data","score":0.3720000088214874},{"id":"https://openalex.org/keywords/bridge","display_name":"Bridge (graph theory)","score":0.36079999804496765},{"id":"https://openalex.org/keywords/model-selection","display_name":"Model selection","score":0.33469998836517334},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.3025999963283539},{"id":"https://openalex.org/keywords/raw-data","display_name":"Raw data","score":0.2955000102519989}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.804099977016449},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.7889000177383423},{"id":"https://openalex.org/C2992525071","wikidata":"https://www.wikidata.org/wiki/Q50818671","display_name":"Federated learning","level":2,"score":0.5627999901771545},{"id":"https://openalex.org/C1893757","wikidata":"https://www.wikidata.org/wiki/Q3653001","display_name":"Inversion (geology)","level":3,"score":0.5080000162124634},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.42969998717308044},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3813999891281128},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3727000057697296},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.3720000088214874},{"id":"https://openalex.org/C100776233","wikidata":"https://www.wikidata.org/wiki/Q2532492","display_name":"Bridge (graph theory)","level":2,"score":0.36079999804496765},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.35120001435279846},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.337799996137619},{"id":"https://openalex.org/C93959086","wikidata":"https://www.wikidata.org/wiki/Q6888345","display_name":"Model selection","level":2,"score":0.33469998836517334},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.3025999963283539},{"id":"https://openalex.org/C132964779","wikidata":"https://www.wikidata.org/wiki/Q2110223","display_name":"Raw data","level":2,"score":0.2955000102519989},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.29010000824928284},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.2858000099658966},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.27970001101493835},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.2782000005245209},{"id":"https://openalex.org/C2776145971","wikidata":"https://www.wikidata.org/wiki/Q30673951","display_name":"Labeled data","level":2,"score":0.27790001034736633},{"id":"https://openalex.org/C77553402","wikidata":"https://www.wikidata.org/wiki/Q13222579","display_name":"Upper and lower bounds","level":2,"score":0.26910001039505005},{"id":"https://openalex.org/C123201435","wikidata":"https://www.wikidata.org/wiki/Q456632","display_name":"Information privacy","level":2,"score":0.26750001311302185},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.2651999890804291},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.26489999890327454},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.263700008392334},{"id":"https://openalex.org/C35525427","wikidata":"https://www.wikidata.org/wiki/Q745881","display_name":"Intrusion detection system","level":2,"score":0.2599000036716461},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.25429999828338623},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.2533999979496002},{"id":"https://openalex.org/C17020691","wikidata":"https://www.wikidata.org/wiki/Q139677","display_name":"Operator (biology)","level":5,"score":0.2515999972820282}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.17623","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17623","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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.17623","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17623","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":"article"},"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","score":0.5348897576332092,"id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Federated":[0],"Learning":[1],"(FL)":[2],"enables":[3],"collaborative":[4],"model":[5,9],"training":[6,32,92,96],"by":[7,76],"sharing":[8],"updates":[10,27],"instead":[11],"of":[12,49],"raw":[13],"data,":[14],"aiming":[15],"to":[16,90,131,139],"protect":[17],"user":[18],"privacy.":[19],"However,":[20],"recent":[21],"studies":[22],"reveal":[23],"that":[24,176,199],"these":[25],"shared":[26],"can":[28],"inadvertently":[29],"leak":[30],"sensitive":[31],"data":[33],"through":[34],"gradient":[35],"inversion":[36,83],"attacks":[37],"(GIAs).":[38],"Among":[39],"them,":[40],"active":[41,87],"GIAs":[42,187],"are":[43],"particularly":[44],"powerful,":[45],"enabling":[46,142],"high-fidelity":[47,179],"reconstruction":[48,160,180],"individual":[50],"samples":[51,93],"even":[52],"under":[53,188],"large":[54,95,189],"batch":[55,190],"sizes.":[56],"Nevertheless,":[57],"existing":[58],"approaches":[59],"often":[60],"require":[61],"architectural":[62,100],"modifications,":[63],"which":[64],"limit":[65],"their":[66],"practical":[67],"applicability.":[68],"In":[69],"this":[70,74],"work,":[71],"we":[72,103],"bridge":[73],"gap":[75],"introducing":[77],"the":[78,105,118,129,136,149,159,166,212],"Activation":[79],"REcovery":[80],"via":[81],"Sparse":[82],"(ARES)":[84],"attack,":[85],"an":[86,155],"GIA":[88],"designed":[89],"reconstruct":[91],"from":[94],"batches":[97],"without":[98],"requiring":[99],"modifications.":[101],"Specifically,":[102],"formulate":[104],"recovery":[106,112,151],"problem":[107],"as":[108],"a":[109,203],"noisy":[110],"sparse":[111],"task":[113],"and":[114,123,153,173,192,205],"solve":[115],"it":[116],"using":[117],"generalized":[119],"Least":[120],"Absolute":[121],"Shrinkage":[122],"Selection":[124],"Operator":[125],"(Lasso).":[126],"To":[127],"extend":[128],"attack":[130],"multi-sample":[132],"recovery,":[133],"ARES":[134,167,177],"incorporates":[135],"imprint":[137],"method":[138],"disentangle":[140],"activations,":[141],"scalable":[143],"per-sample":[144],"reconstruction.":[145],"We":[146],"further":[147],"establish":[148],"expected":[150],"rate":[152],"derive":[154],"upper":[156],"bound":[157],"on":[158,171],"error,":[161],"providing":[162],"theoretical":[163],"guarantees":[164],"for":[165,215],"attack.":[168],"Extensive":[169],"experiments":[170],"CNNs":[172],"MLPs":[174],"demonstrate":[175],"achieves":[178],"across":[181],"diverse":[182],"datasets,":[183],"significantly":[184],"outperforming":[185],"prior":[186],"sizes":[191],"realistic":[193],"FL":[194],"settings.":[195],"Our":[196],"results":[197],"highlight":[198],"intermediate":[200],"activations":[201],"pose":[202],"serious":[204],"underestimated":[206],"privacy":[207],"risk":[208],"in":[209],"FL,":[210],"underscoring":[211],"urgent":[213],"need":[214],"stronger":[216],"defenses.":[217]},"counts_by_year":[],"updated_date":"2026-03-20T20:54:20.808490","created_date":"2026-03-20T00:00:00"}