{"id":"https://openalex.org/W4415540724","doi":"https://doi.org/10.1145/3746027.3755142","title":"MiraGe: Multimodal Discriminative Representation Learning for Generalizable AI-Generated Image Detection","display_name":"MiraGe: Multimodal Discriminative Representation Learning for Generalizable AI-Generated Image Detection","publication_year":2025,"publication_date":"2025-10-25","ids":{"openalex":"https://openalex.org/W4415540724","doi":"https://doi.org/10.1145/3746027.3755142"},"language":null,"primary_location":{"id":"doi:10.1145/3746027.3755142","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746027.3755142","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 33rd ACM International Conference on Multimedia","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3746027.3755142","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5107808169","display_name":"Kuo Shi","orcid":"https://orcid.org/0009-0009-2753-2270"},"institutions":[{"id":"https://openalex.org/I114017466","display_name":"University of Technology Sydney","ror":"https://ror.org/03f0f6041","country_code":"AU","type":"education","lineage":["https://openalex.org/I114017466"]}],"countries":["AU"],"is_corresponding":true,"raw_author_name":"Kuo Shi","raw_affiliation_strings":["University of Technology Sydney, Ultimo, NSW, Australia"],"affiliations":[{"raw_affiliation_string":"University of Technology Sydney, Ultimo, NSW, Australia","institution_ids":["https://openalex.org/I114017466"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100675577","display_name":"Jie L\u00fc","orcid":"https://orcid.org/0000-0003-0690-4732"},"institutions":[{"id":"https://openalex.org/I114017466","display_name":"University of Technology Sydney","ror":"https://ror.org/03f0f6041","country_code":"AU","type":"education","lineage":["https://openalex.org/I114017466"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Jie Lu","raw_affiliation_strings":["University of Technology Sydney, Ultimo, NSW, Australia"],"affiliations":[{"raw_affiliation_string":"University of Technology Sydney, Ultimo, NSW, Australia","institution_ids":["https://openalex.org/I114017466"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056384174","display_name":"Shanshan Ye","orcid":"https://orcid.org/0009-0003-6961-7455"},"institutions":[{"id":"https://openalex.org/I114017466","display_name":"University of Technology Sydney","ror":"https://ror.org/03f0f6041","country_code":"AU","type":"education","lineage":["https://openalex.org/I114017466"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Shanshan Ye","raw_affiliation_strings":["University of Technology Sydney, Ultimo, NSW, Australia"],"affiliations":[{"raw_affiliation_string":"University of Technology Sydney, Ultimo, NSW, Australia","institution_ids":["https://openalex.org/I114017466"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062192039","display_name":"Guangquan Zhang","orcid":"https://orcid.org/0000-0003-3960-0583"},"institutions":[{"id":"https://openalex.org/I114017466","display_name":"University of Technology Sydney","ror":"https://ror.org/03f0f6041","country_code":"AU","type":"education","lineage":["https://openalex.org/I114017466"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Guangquan Zhang","raw_affiliation_strings":["University of Technology Sydney, Ultimo, NSW, Australia"],"affiliations":[{"raw_affiliation_string":"University of Technology Sydney, Ultimo, NSW, Australia","institution_ids":["https://openalex.org/I114017466"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5087852871","display_name":"Zhen Fang","orcid":"https://orcid.org/0000-0003-0602-6255"},"institutions":[{"id":"https://openalex.org/I114017466","display_name":"University of Technology Sydney","ror":"https://ror.org/03f0f6041","country_code":"AU","type":"education","lineage":["https://openalex.org/I114017466"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Zhen Fang","raw_affiliation_strings":["University of Technology Sydney, Ultimo, NSW, Australia"],"affiliations":[{"raw_affiliation_string":"University of Technology Sydney, Ultimo, NSW, Australia","institution_ids":["https://openalex.org/I114017466"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5107808169"],"corresponding_institution_ids":["https://openalex.org/I114017466"],"apc_list":null,"apc_paid":null,"fwci":2.3568,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.91578259,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"353","last_page":"361"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9983999729156494,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9983999729156494,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9973000288009644,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9965999722480774,"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/discriminative-model","display_name":"Discriminative model","score":0.8274999856948853},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.6988999843597412},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.6040999889373779},{"id":"https://openalex.org/keywords/feature-learning","display_name":"Feature learning","score":0.6035000085830688},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.5217000246047974},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.489300012588501},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.45260000228881836},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.4462999999523163}],"concepts":[{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.8274999856948853},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7706999778747559},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.7494000196456909},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6988999843597412},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.6040999889373779},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.6035000085830688},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.5217000246047974},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5008000135421753},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.489300012588501},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.45260000228881836},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4462999999523163},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.4032000005245209},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.4004000127315521},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.3644999861717224},{"id":"https://openalex.org/C2778334786","wikidata":"https://www.wikidata.org/wiki/Q1586270","display_name":"Variation (astronomy)","level":2,"score":0.34860000014305115},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.31150001287460327},{"id":"https://openalex.org/C2781122975","wikidata":"https://www.wikidata.org/wiki/Q16928266","display_name":"Semantic feature","level":2,"score":0.3001999855041504},{"id":"https://openalex.org/C75294576","wikidata":"https://www.wikidata.org/wiki/Q5165192","display_name":"Contextual image classification","level":3,"score":0.2937999963760376},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2906000018119812},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.289000004529953},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.28450000286102295},{"id":"https://openalex.org/C136389625","wikidata":"https://www.wikidata.org/wiki/Q334384","display_name":"Supervised learning","level":3,"score":0.2623000144958496}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3746027.3755142","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746027.3755142","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 33rd ACM International Conference on Multimedia","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3746027.3755142","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746027.3755142","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 33rd ACM International Conference on Multimedia","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G6774349052","display_name":null,"funder_award_id":"FL190100149","funder_id":"https://openalex.org/F4320334704","funder_display_name":"Australian Research Council"}],"funders":[{"id":"https://openalex.org/F4320334704","display_name":"Australian Research Council","ror":"https://ror.org/05mmh0f86"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":6,"referenced_works":["https://openalex.org/W2007339694","https://openalex.org/W3198377975","https://openalex.org/W4327662657","https://openalex.org/W4392172801","https://openalex.org/W4394862984","https://openalex.org/W4405021796"],"related_works":[],"abstract_inverted_index":{"Recent":[0],"advances":[1],"in":[2],"generative":[3,39],"models":[4,40],"have":[5],"highlighted":[6],"the":[7,89],"need":[8],"for":[9,60,119],"robust":[10],"detectors":[11],"capable":[12],"of":[13],"distinguishing":[14],"real":[15],"images":[16],"from":[17],"AI-generated":[18,62],"images.":[19],"While":[20],"existing":[21],"methods":[22],"perform":[23],"well":[24],"on":[25,77],"known":[26],"generators,":[27],"their":[28],"performance":[29],"often":[30],"declines":[31],"when":[32],"tested":[33],"with":[34],"newly":[35],"emerging":[36],"or":[37],"unseen":[38,142],"due":[41],"to":[42,69,106],"overlapping":[43],"feature":[44,98],"embeddings":[45,115],"that":[46,133],"hinder":[47],"accurate":[48],"cross-generator":[49],"classification.":[50],"In":[51],"this":[52],"paper,":[53],"we":[54,101],"propose":[55],"Multimodal":[56],"Discriminative":[57],"Representation":[58],"Learning":[59],"Generalizable":[61],"Image":[63],"Detection":[64],"(MiraGe),":[65],"a":[66],"method":[67],"designed":[68],"learn":[70],"generator-invariant":[71],"features.":[72],"Motivated":[73],"by":[74],"theoretical":[75],"insights":[76],"intra-class":[78],"variation":[79],"minimization":[80],"and":[81],"inter-class":[82],"separation,":[83],"MiraGe":[84,134],"tightly":[85],"aligns":[86],"features":[87],"within":[88],"same":[90],"class":[91],"while":[92],"maximizing":[93],"separation":[94],"between":[95],"classes,":[96],"enhancing":[97],"discriminability.":[99],"Moreover,":[100],"apply":[102],"multimodal":[103],"prompt":[104],"learning":[105],"further":[107],"refine":[108],"these":[109],"principles":[110],"into":[111],"CLIP,":[112],"leveraging":[113],"text":[114],"as":[116],"semantic":[117],"anchors":[118],"effective":[120],"discriminative":[121],"representation":[122],"learning,":[123],"thereby":[124],"improving":[125],"generalizability.":[126],"Comprehensive":[127],"experiments":[128],"across":[129],"multiple":[130],"benchmarks":[131],"show":[132],"achieves":[135],"state-of-the-art":[136],"performance,":[137],"maintaining":[138],"robustness":[139],"even":[140],"against":[141],"generators":[143],"like":[144],"Sora.":[145]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-04-09T08:11:56.329763","created_date":"2025-10-25T00:00:00"}