{"id":"https://openalex.org/W7154468971","doi":"https://doi.org/10.48550/arxiv.2604.12353","title":"Combating Pattern and Content Bias: Adversarial Feature Learning for Generalized AI-Generated Image Detection","display_name":"Combating Pattern and Content Bias: Adversarial Feature Learning for Generalized AI-Generated Image Detection","publication_year":2026,"publication_date":"2026-04-14","ids":{"openalex":"https://openalex.org/W7154468971","doi":"https://doi.org/10.48550/arxiv.2604.12353"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.12353","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.12353","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.2604.12353","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5133641201","display_name":"Haifeng Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Haifeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133636933","display_name":"Qinghui He","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Qinghui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133686420","display_name":"Xiuli Bi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bi, Xiuli","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133709516","display_name":"Bo Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Bo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133667103","display_name":"Chi-Man Pun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pun, Chi-Man","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5133639624","display_name":"Bin Xiao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiao, Bin","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":0.6819000244140625,"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":0.6819000244140625,"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/T11147","display_name":"Misinformation and Its Impacts","score":0.15000000596046448,"subfield":{"id":"https://openalex.org/subfields/3312","display_name":"Sociology and Political Science"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.03290000185370445,"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/adversarial-system","display_name":"Adversarial system","score":0.7926999926567078},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.6747999787330627},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.670799970626831},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.652400016784668},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.5871999859809875},{"id":"https://openalex.org/keywords/feature-learning","display_name":"Feature learning","score":0.534500002861023},{"id":"https://openalex.org/keywords/feature-extraction","display_name":"Feature extraction","score":0.5325000286102295},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.4867999851703644},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.4569000005722046}],"concepts":[{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.7926999926567078},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.736299991607666},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.7218000292778015},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.6747999787330627},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.670799970626831},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.652400016784668},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.5871999859809875},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.534500002861023},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.5325000286102295},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5001000165939331},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.4867999851703644},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4569000005722046},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.4359000027179718},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.3962000012397766},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.3921999931335449},{"id":"https://openalex.org/C83665646","wikidata":"https://www.wikidata.org/wiki/Q42139305","display_name":"Feature vector","level":2,"score":0.3296000063419342},{"id":"https://openalex.org/C101738243","wikidata":"https://www.wikidata.org/wiki/Q786435","display_name":"Autoencoder","level":3,"score":0.32820001244544983},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.31949999928474426},{"id":"https://openalex.org/C2988773926","wikidata":"https://www.wikidata.org/wiki/Q25104379","display_name":"Generative adversarial network","level":3,"score":0.28760001063346863},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.2872999906539917},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.26930001378059387},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2651999890804291},{"id":"https://openalex.org/C75294576","wikidata":"https://www.wikidata.org/wiki/Q5165192","display_name":"Contextual image classification","level":3,"score":0.25600001215934753},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.25189998745918274}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.12353","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.12353","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.2604.12353","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.12353","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/10","display_name":"Reduced inequalities","score":0.6908093690872192}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"In":[0],"recent":[1],"years,":[2],"the":[3,14,54,69,104,147,152,163,177],"rapid":[4],"development":[5],"of":[6],"generative":[7,63,77,153,158],"artificial":[8],"intelligence":[9],"technology":[10],"has":[11],"significantly":[12],"lowered":[13],"barrier":[15],"to":[16,25,50,60,149],"creating":[17],"high-quality":[18],"fake":[19],"images,":[20,170,210],"posing":[21],"a":[22,88,98,109,122],"serious":[23],"challenge":[24],"information":[26],"authenticity":[27],"and":[28,65,114,135,142,168,174,197],"credibility.":[29],"Existing":[30],"generated":[31,169],"image":[32,101],"detection":[33,217],"methods":[34],"typically":[35],"enhance":[36],"generalization":[37,46],"through":[38],"model":[39,148],"architecture":[40],"or":[41],"network":[42],"design.":[43],"However,":[44],"their":[45],"performance":[47],"remains":[48],"susceptible":[49],"data":[51,56],"bias,":[52],"as":[53,103],"training":[55,129,181],"may":[57],"drive":[58],"models":[59,78],"fit":[61],"specific":[62],"patterns":[64],"content":[66,143],"rather":[67],"than":[68],"common":[70],"features":[71,154],"shared":[72,155],"by":[73,193],"images":[74],"from":[75],"different":[76,157],"(asymmetric":[79],"bias":[80,136],"learning).":[81],"To":[82],"address":[83],"this":[84],"issue,":[85],"we":[86],"propose":[87],"Multi-dimensional":[89],"Adversarial":[90],"Feature":[91],"Learning":[92],"(MAFL)":[93],"framework.":[94],"The":[95],"framework":[96],"adopts":[97],"pretrained":[99],"multimodal":[100],"encoder":[102],"feature":[105,111,133,137],"extraction":[106],"backbone,":[107],"constructs":[108],"real-fake":[110],"learning":[112,134],"network,":[113],"designs":[115],"an":[116,127],"adversarial":[117,124,128],"bias-learning":[118],"branch":[119],"equipped":[120],"with":[121,207],"multi-dimensional":[123],"loss,":[125],"forming":[126],"mechanism":[130],"between":[131,166],"authenticity-discriminative":[132],"learning.":[138],"By":[139],"suppressing":[140],"generation-pattern":[141],"biases,":[144],"MAFL":[145],"guides":[146],"focus":[150],"on":[151,179,219],"across":[156],"models,":[159],"thereby":[160],"effectively":[161],"capturing":[162],"fundamental":[164],"differences":[165],"real":[167],"enhancing":[171],"cross-model":[172],"generalization,":[173],"substantially":[175],"reducing":[176],"reliance":[178],"large-scale":[180],"data.":[182],"Through":[183],"extensive":[184],"experimental":[185],"validation,":[186],"our":[187],"method":[188],"outperforms":[189],"existing":[190],"state-of-the-art":[191],"approaches":[192],"10.89%":[194],"in":[195,199],"accuracy":[196,218],"8.57%":[198],"Average":[200],"Precision":[201],"(AP).":[202],"Notably,":[203],"even":[204],"when":[205],"trained":[206],"only":[208],"320":[209],"it":[211],"can":[212],"still":[213],"achieve":[214],"over":[215],"80%":[216],"public":[220],"datasets.":[221]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-16T00:00:00"}