{"id":"https://openalex.org/W7171695138","doi":"https://doi.org/10.48550/arxiv.2607.25502","title":"Anti-Backdoor Coreset Selection via Cumulative Entropy","display_name":"Anti-Backdoor Coreset Selection via Cumulative Entropy","publication_year":2026,"publication_date":"2026-07-28","ids":{"openalex":"https://openalex.org/W7171695138","doi":"https://doi.org/10.48550/arxiv.2607.25502"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2607.25502","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.25502","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":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.2607.25502","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5143891909","display_name":"Qi Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Qi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5040751355","display_name":"Christian Wressnegger","orcid":"https://orcid.org/0009-0007-1493-9552"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wressnegger, Christian","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":0.946399986743927,"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":0.946399986743927,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.014999999664723873,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.006200000178068876,"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/backdoor","display_name":"Backdoor","score":0.5745999813079834},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.5198000073432922},{"id":"https://openalex.org/keywords/entropy","display_name":"Entropy (arrow of time)","score":0.4415999948978424},{"id":"https://openalex.org/keywords/principle-of-maximum-entropy","display_name":"Principle of maximum entropy","score":0.4196000099182129},{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.4163999855518341},{"id":"https://openalex.org/keywords/model-selection","display_name":"Model selection","score":0.36230000853538513}],"concepts":[{"id":"https://openalex.org/C2781045450","wikidata":"https://www.wikidata.org/wiki/Q254569","display_name":"Backdoor","level":2,"score":0.5745999813079834},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5648999810218811},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5453000068664551},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.5198000073432922},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.517300009727478},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.4415999948978424},{"id":"https://openalex.org/C9679016","wikidata":"https://www.wikidata.org/wiki/Q1417473","display_name":"Principle of maximum entropy","level":2,"score":0.4196000099182129},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.4163999855518341},{"id":"https://openalex.org/C93959086","wikidata":"https://www.wikidata.org/wiki/Q6888345","display_name":"Model selection","level":2,"score":0.36230000853538513},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.35339999198913574},{"id":"https://openalex.org/C148483581","wikidata":"https://www.wikidata.org/wiki/Q446488","display_name":"Feature selection","level":2,"score":0.33889999985694885},{"id":"https://openalex.org/C75268714","wikidata":"https://www.wikidata.org/wiki/Q43478","display_name":"Natural selection","level":3,"score":0.33719998598098755},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.30880001187324524},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.2806999981403351},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2791000008583069}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2607.25502","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.25502","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":"doi:10.48550/arxiv.2607.25502","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.25502","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":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recent":[0],"training-time":[1,131],"defenses":[2,145],"against":[3,152],"neural":[4],"backdoors":[5],"isolate":[6],"a":[7,16,30,135,140,162],"benign":[8,55,66,99,121,136],"subset":[9],"from":[10,19],"poisoned":[11],"training":[12,92],"data,":[13],"to":[14,36,80,97,116,138],"learn":[15],"backdoor-free":[17,141],"model":[18],"it.":[20],"In":[21],"this":[22,26,83,126],"paper,":[23],"we":[24,108],"formulate":[25],"defense":[27,132],"strategy":[28],"as":[29,77],"coreset":[31,57,137],"selection":[32,58,78],"problem,":[33],"giving":[34],"rise":[35],"so-called":[37],"\"Anti-Backdoor":[38],"Coreset":[39],"Selection.\"":[40],"Since":[41],"poisonous":[42,123],"samples":[43,63,93,100,112],"have":[44],"(a)":[45],"lower":[46],"prediction":[47],"uncertainty":[48],"and":[49,94,122,150],"are":[50],"(b)":[51],"less":[52],"frequent":[53],"than":[54,68],"samples,":[56],"naturally":[59],"focuses":[60],"more":[61],"on":[62,165],"associated":[64],"with":[65,101,161],"functionality":[67],"the":[69,74,88,105,110,118],"backdoor":[70],"functionality.":[71],"We":[72],"use":[73],"Cumulative":[75],"Entropy":[76],"criterion":[79],"further":[81],"facilitate":[82,117],"effect.":[84],"The":[85],"metric":[86],"tracks":[87],"learning":[89],"dynamics":[90],"of":[91],"allowing":[95],"us":[96],"select":[98],"high":[102],"informativeness":[103],"for":[104],"coreset.":[106],"Additionally,":[107],"unlearn":[109],"chosen":[111],"in":[113],"each":[114],"epoch":[115],"separability":[119],"between":[120],"samples.":[124],"Together,":[125],"yields":[127],"an":[128],"exceptionally":[129],"effective":[130],"that":[133,146],"constructs":[134],"train":[139],"model.":[142],"Unlike":[143],"prior":[144],"compromise":[147],"natural":[148,166],"accuracy":[149],"fail":[151],"certain":[153],"attacks,":[154],"our":[155],"method":[156],"mitigates":[157],"backdooring":[158],"attacks":[159],"consistently":[160],"negligible":[163],"impact":[164],"performance.":[167]},"counts_by_year":[],"updated_date":"2026-07-30T17:31:21.811387","created_date":"2026-07-30T00:00:00"}