{"id":"https://openalex.org/W7140206851","doi":"https://doi.org/10.48550/arxiv.2603.21939","title":"FeatDistill: A Feature Distillation Enhanced Multi-Expert Ensemble Framework for Robust AI-generated Image Detection","display_name":"FeatDistill: A Feature Distillation Enhanced Multi-Expert Ensemble Framework for Robust AI-generated Image Detection","publication_year":2026,"publication_date":"2026-03-23","ids":{"openalex":"https://openalex.org/W7140206851","doi":"https://doi.org/10.48550/arxiv.2603.21939"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.21939","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.21939","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":"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.2603.21939","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Tu, Zhilin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tu, Zhilin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Li, Kemou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Kemou","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Li, Fengpeng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Fengpeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Fei, Jianwei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fei, Jianwei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Zhang, Jiamin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Jiamin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"Wu, Haiwei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Haiwei","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/T12357","display_name":"Digital Media Forensic Detection","score":0.8634999990463257,"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"}},"topics":[{"id":"https://openalex.org/T12357","display_name":"Digital Media Forensic Detection","score":0.8634999990463257,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.04820000007748604,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.03530000150203705,"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/overfitting","display_name":"Overfitting","score":0.7562000155448914},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.5927000045776367},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4684999883174896},{"id":"https://openalex.org/keywords/ensemble-learning","display_name":"Ensemble learning","score":0.4472000002861023},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.4410000145435333},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.4325000047683716},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.40639999508857727},{"id":"https://openalex.org/keywords/object-detection","display_name":"Object detection","score":0.39629998803138733},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.3774000108242035}],"concepts":[{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.7562000155448914},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.753600001335144},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6298999786376953},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.5927000045776367},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5327000021934509},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4684999883174896},{"id":"https://openalex.org/C45942800","wikidata":"https://www.wikidata.org/wiki/Q245652","display_name":"Ensemble learning","level":2,"score":0.4472000002861023},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.4410000145435333},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4325000047683716},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.40639999508857727},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.39629998803138733},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.39399999380111694},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.3774000108242035},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.3564000129699707},{"id":"https://openalex.org/C75294576","wikidata":"https://www.wikidata.org/wiki/Q5165192","display_name":"Contextual image classification","level":3,"score":0.3538999855518341},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.3458999991416931},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.3409999907016754},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.32589998841285706},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.32330000400543213},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.31690001487731934},{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.31529998779296875},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.31209999322891235},{"id":"https://openalex.org/C94915269","wikidata":"https://www.wikidata.org/wiki/Q1834857","display_name":"Detector","level":2,"score":0.28529998660087585},{"id":"https://openalex.org/C66905080","wikidata":"https://www.wikidata.org/wiki/Q17005494","display_name":"Binary classification","level":3,"score":0.28299999237060547},{"id":"https://openalex.org/C2776836416","wikidata":"https://www.wikidata.org/wiki/Q1364844","display_name":"False alarm","level":2,"score":0.2786000072956085},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.2736000120639801},{"id":"https://openalex.org/C739882","wikidata":"https://www.wikidata.org/wiki/Q3560506","display_name":"Anomaly detection","level":2,"score":0.2578999996185303}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.21939","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.21939","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":"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.2603.21939","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.21939","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":"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":[{"id":"https://metadata.un.org/sdg/16","score":0.6563605666160583,"display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0,59],"rapid":[1],"iteration":[2],"and":[3,18,74,89,106,119,159,185,191,223,232],"widespread":[4],"dissemination":[5],"of":[6,21,87,163],"deepfake":[7,237],"technology":[8],"have":[9],"posed":[10],"severe":[11],"challenges":[12],"to":[13,92,115],"information":[14],"security,":[15],"making":[16],"robust":[17],"generalizable":[19],"detection":[20,36],"AI-generated":[22,34],"forged":[23],"images":[24],"increasingly":[25],"important.":[26],"In":[27],"this":[28],"paper,":[29],"we":[30,78,101],"propose":[31],"FeatDistill,":[32],"an":[33,230],"image":[35,238],"framework":[37,60,199],"that":[38,218],"integrates":[39],"feature":[40,72],"distillation":[41],"with":[42,139],"a":[43,80,130,140],"multi-expert":[44],"ensemble,":[45],"developed":[46],"for":[47,151,235],"the":[48,57,103,113,134,169,176,195,198,213],"NTIRE":[49,214],"Challenge":[50],"on":[51],"Robust":[52],"AI-Generated":[53],"Image":[54],"Detection":[55],"in":[56,66,124,212],"Wild.":[58],"explicitly":[61],"targets":[62],"three":[63],"practical":[64,233],"bottlenecks":[65],"real-world":[67,236],"forensics:":[68],"degradation":[69,109],"interference,":[70],"insufficient":[71],"representation,":[73],"limited":[75],"generalization.":[76],"Concretely,":[77],"build":[79],"four-backbone":[81],"Vision":[82],"Transformer":[83],"(ViT)":[84],"ensemble":[85,196],"composed":[86],"CLIP":[88],"SigLIP":[90],"variants":[91],"capture":[93],"complementary":[94],"forensic":[95],"cues.":[96],"To":[97],"improve":[98],"data":[99],"coverage,":[100],"expand":[102],"training":[104,132],"set":[105],"introduce":[107],"comprehensive":[108],"modeling,":[110],"which":[111],"exposes":[112],"detector":[114],"diverse":[116,226],"quality":[117],"variations":[118],"synthesis":[120],"artifacts":[121],"commonly":[122],"encountered":[123],"unconstrained":[125],"scenarios.":[126],"We":[127],"further":[128],"adopt":[129],"two-stage":[131],"paradigm:":[133],"model":[135],"is":[136,172],"first":[137],"optimized":[138],"standard":[141],"binary":[142],"classification":[143],"objective,":[144],"then":[145],"refined":[146],"by":[147,174],"dense":[148],"feature-level":[149],"self-distillation":[150],"representation":[152],"alignment.":[153],"This":[154],"design":[155],"effectively":[156],"mitigates":[157],"overfitting":[158],"enhances":[160],"semantic":[161],"consistency":[162],"learned":[164],"features.":[165],"At":[166],"inference":[167],"time,":[168],"final":[170],"prediction":[171],"obtained":[173],"averaging":[175],"probabilities":[177],"from":[178],"four":[179],"independently":[180],"trained":[181],"experts,":[182],"yielding":[183],"stable":[184],"reliable":[186],"decisions":[187],"across":[188],"unseen":[189],"generators":[190],"complex":[192],"degradations.":[193],"Despite":[194],"design,":[197],"remains":[200],"efficient,":[201],"requiring":[202],"only":[203],"about":[204],"10":[205],"GB":[206],"peak":[207],"GPU":[208],"memory.":[209],"Extensive":[210],"evaluations":[211],"challenge":[215],"setting":[216],"demonstrate":[217],"FeatDistill":[219],"achieves":[220],"strong":[221],"robustness":[222],"generalization":[224],"under":[225],"``in-the-wild''":[227],"conditions,":[228],"offering":[229],"effective":[231],"solution":[234],"detection.":[239]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-03-25T00:00:00"}