{"id":"https://openalex.org/W7133351996","doi":"https://doi.org/10.48550/arxiv.2603.01450","title":"Deepfake Forensics Adapter: A Dual-Stream Network for Generalizable Deepfake Detection","display_name":"Deepfake Forensics Adapter: A Dual-Stream Network for Generalizable Deepfake Detection","publication_year":2026,"publication_date":"2026-03-02","ids":{"openalex":"https://openalex.org/W7133351996","doi":"https://doi.org/10.48550/arxiv.2603.01450"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.01450","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.01450","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.2603.01450","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5127885396","display_name":"Jianfeng Liao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liao, Jianfeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127904159","display_name":"Yichen Wei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei, Yichen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127927491","display_name":"Raymond Chan Ching Bon","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bon, Raymond Chan Ching","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127978815","display_name":"Shulan Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Shulan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102173312","display_name":"Kam-Pui Chow","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chow, Kam-Pui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5127974146","display_name":"Kwok-Yan Lam","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lam, Kwok-Yan","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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.817300021648407,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.817300021648407,"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/T12357","display_name":"Digital Media Forensic Detection","score":0.08429999649524689,"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/T11448","display_name":"Face recognition and analysis","score":0.028300000354647636,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.43939998745918274},{"id":"https://openalex.org/keywords/classifier","display_name":"Classifier (UML)","score":0.4162999987602234},{"id":"https://openalex.org/keywords/anomaly-detection","display_name":"Anomaly detection","score":0.3547999858856201},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.3343000113964081},{"id":"https://openalex.org/keywords/object-detection","display_name":"Object detection","score":0.3325999975204468},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.32280001044273376}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8245000243186951},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5548999905586243},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.43939998745918274},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4375},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.4162999987602234},{"id":"https://openalex.org/C739882","wikidata":"https://www.wikidata.org/wiki/Q3560506","display_name":"Anomaly detection","level":2,"score":0.3547999858856201},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.3343000113964081},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.3325999975204468},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.32280001044273376},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.31869998574256897},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3154999911785126},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.3075999915599823},{"id":"https://openalex.org/C177284502","wikidata":"https://www.wikidata.org/wiki/Q1005390","display_name":"Adapter (computing)","level":2,"score":0.30469998717308044},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.2590999901294708}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.01450","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.01450","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.2603.01450","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.01450","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":[{"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16","score":0.7368594408035278}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0],"rapid":[1],"advancement":[2],"of":[3,20,80,147,157,171,176],"deepfake":[4,72,207,217],"generation":[5],"techniques":[6],"poses":[7],"significant":[8],"threats":[9],"to":[10,34,69,93,113],"public":[11],"safety":[12],"and":[13,125,134,138,149,173,200],"causes":[14],"societal":[15],"harm":[16],"through":[17],"the":[18,76,110,153,164,215],"creation":[19],"highly":[21],"realistic":[22],"synthetic":[23],"facial":[24,116,122],"media.":[25],"While":[26],"existing":[27],"detection":[28,73,208],"methods":[29],"demonstrate":[30,152],"limitations":[31],"in":[32,97,163],"generalizing":[33],"emerging":[35],"forgery":[36,117],"patterns,":[37],"this":[38],"paper":[39],"presents":[40],"Deepfake":[41],"Forensics":[42],"Adapter":[43,90],"(DFA),":[44],"a":[45,61,142,179,198,205],"novel":[46],"dual-stream":[47],"framework":[48,188],"that":[49,100],"synergizes":[50],"vision-language":[51],"foundation":[52],"models":[53],"with":[54,65,168,210],"targeted":[55],"forensics":[56],"analysis.":[57],"Our":[58,187,219],"approach":[59],"integrates":[60],"pre-trained":[62],"CLIP":[63,81,84],"model":[64],"three":[66],"core":[67],"components":[68],"achieve":[70],"specialized":[71],"by":[74,119],"leveraging":[75,121],"powerful":[77],"general":[78],"capabilities":[79,156,213],"without":[82],"changing":[83],"parameters:":[85],"1)":[86],"A":[87,105],"Global":[88],"Feature":[89],"is":[91,221],"used":[92],"identify":[94],"global":[95,137],"inconsistencies":[96],"image":[98],"content":[99],"may":[101],"indicate":[102],"forgery,":[103],"2)":[104],"Local":[106],"Anomaly":[107],"Stream":[108],"enhances":[109],"model's":[111],"ability":[112],"perceive":[114],"local":[115,139],"cues":[118],"explicitly":[120],"structure":[123],"priors,":[124],"3)":[126],"An":[127],"Interactive":[128],"Fusion":[129],"Classifier":[130],"promotes":[131],"deep":[132],"interaction":[133],"fusion":[135],"between":[136],"features":[140],"using":[141],"transformer":[143],"encoder.":[144],"Extensive":[145],"evaluations":[146],"frame-level":[148,169],"video-level":[150,174],"benchmarks":[151],"superior":[154],"generalization":[155,212],"DFA,":[158],"particularly":[159],"achieving":[160],"state-of-the-art":[161,192],"performance":[162],"challenging":[165],"DFDC":[166],"dataset":[167],"AUC/EER":[170,175],"0.816/0.256":[172],"0.836/0.251,":[177],"representing":[178],"4.8%":[180],"video":[181],"AUC":[182],"improvement":[183],"over":[184],"previous":[185],"methods.":[186],"not":[189],"only":[190],"demonstrates":[191],"performance,":[193],"but":[194],"also":[195],"points":[196],"out":[197],"feasible":[199],"effective":[201],"direction":[202],"for":[203],"developing":[204],"robust":[206],"system":[209],"enhanced":[211],"against":[214],"evolving":[216],"threats.":[218],"code":[220],"available":[222],"at":[223],"https://github.com/Liao330/DFA.git":[224]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-03-04T00:00:00"}