{"id":"https://openalex.org/W7166709281","doi":"https://doi.org/10.48550/arxiv.2606.30528","title":"$\u03bc$Flow: Leveraging Average Images for Improving Generalisation of Deepfake Faces Detectors","display_name":"$\u03bc$Flow: Leveraging Average Images for Improving Generalisation of Deepfake Faces Detectors","publication_year":2026,"publication_date":"2026-06-29","ids":{"openalex":"https://openalex.org/W7166709281","doi":"https://doi.org/10.48550/arxiv.2606.30528"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.30528","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.30528","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":null,"license_id":null,"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.2606.30528","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5093875947","display_name":"Orazio Pontorno","orcid":"https://orcid.org/0009-0009-0381-9971"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pontorno, Orazio","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019216303","display_name":"Mattia Litrico","orcid":"https://orcid.org/0000-0002-4914-174X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Litrico, Mattia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139677941","display_name":"Luca Guarnera","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guarnera, Luca","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030914810","display_name":"Mario Valerio Giuffrida","orcid":"https://orcid.org/0000-0002-5232-677X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Giuffrida, Mario Valerio","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139632246","display_name":"Sebastiano Battiato","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Battiato, Sebastiano","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":0.9081000089645386,"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":0.9081000089645386,"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":0.03709999844431877,"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":0.02630000002682209,"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":0.7720000147819519},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.5927000045776367},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.5690000057220459},{"id":"https://openalex.org/keywords/detector","display_name":"Detector","score":0.5109999775886536},{"id":"https://openalex.org/keywords/property","display_name":"Property (philosophy)","score":0.5059999823570251},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.492900013923645},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.4867999851703644},{"id":"https://openalex.org/keywords/feature-vector","display_name":"Feature vector","score":0.4634999930858612}],"concepts":[{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.7720000147819519},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7224000096321106},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.684499979019165},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.5927000045776367},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.5690000057220459},{"id":"https://openalex.org/C94915269","wikidata":"https://www.wikidata.org/wiki/Q1834857","display_name":"Detector","level":2,"score":0.5109999775886536},{"id":"https://openalex.org/C189950617","wikidata":"https://www.wikidata.org/wiki/Q937228","display_name":"Property (philosophy)","level":2,"score":0.5059999823570251},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.492900013923645},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.4867999851703644},{"id":"https://openalex.org/C83665646","wikidata":"https://www.wikidata.org/wiki/Q42139305","display_name":"Feature vector","level":2,"score":0.4634999930858612},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.4465000033378601},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3644999861717224},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.34439998865127563},{"id":"https://openalex.org/C2776145971","wikidata":"https://www.wikidata.org/wiki/Q30673951","display_name":"Labeled data","level":2,"score":0.3246999979019165},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3228999972343445},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.30399999022483826},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.29789999127388},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.29010000824928284},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.2678000032901764},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.2655999958515167},{"id":"https://openalex.org/C136389625","wikidata":"https://www.wikidata.org/wiki/Q334384","display_name":"Supervised learning","level":3,"score":0.2637999951839447}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.30528","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.30528","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.30528","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.30528","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"display_name":"Reduced inequalities","score":0.7318720817565918,"id":"https://metadata.un.org/sdg/10"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Current":[0],"generative":[1,94],"models,":[2,7],"including":[3],"GANs":[4,57],"and":[5,20,45,115,140,158],"diffusion":[6],"have":[8],"reached":[9],"an":[10],"outstanding":[11],"level":[12],"of":[13,109,125],"photorealism,":[14],"posing":[15],"significant":[16],"risks":[17],"to":[18,31,120],"privacy":[19],"security.":[21],"To":[22],"ensure":[23],"real-world":[24],"applicability,":[25],"deepfake":[26,68],"detectors":[27],"must":[28],"generalise":[29],"effectively":[30],"unseen":[32,162],"generators.":[33],"However,":[34],"most":[35],"existing":[36],"approaches":[37],"rely":[38],"on":[39,72,77,85,150],"supervised":[40],"training":[41,116],"with":[42,128],"both":[43,156],"real":[44,73,139,157],"fake":[46,141,159],"images,":[47],"which":[48],"limits":[49],"their":[50],"generalisation":[51],"especially":[52],"across":[53],"generators":[54],"categories":[55],"(e.g.":[56],"vs":[58],"DMs).":[59],"In":[60],"this":[61,103,129],"work,":[62],"we":[63],"introduce":[64],"$\u03bc$Flow,":[65],"a":[66,117,134,151],"one-class":[67],"detector":[69],"trained":[70],"only":[71],"images":[74,91,114,127],"without":[75],"relying":[76],"pseudo-deepfakes":[78],"or":[79],"synthetic":[80],"artifacts.":[81],"Our":[82],"approach":[83],"builds":[84],"the":[86,107,122],"observation":[87],"that":[88,137,168],"averaging":[89],"multiple":[90],"amplifies":[92],"consistent":[93],"traces,":[95],"producing":[96],"highly":[97],"discriminative":[98],"feature":[99,123],"representations.":[100],"We":[101,147],"leverage":[102],"property":[104],"by":[105],"modelling":[106],"distribution":[108],"features":[110],"extracted":[111],"from":[112],"averaged":[113],"normalizing":[118],"flow":[119],"align":[121],"space":[124],"individual":[126],"distribution.":[130],"This":[131],"alignment":[132],"yields":[133],"likelihood-based":[135],"criterion":[136],"separates":[138],"samples":[142],"while":[143],"promoting":[144],"strong":[145],"generalisation.":[146],"evaluate":[148],"$\u03bc$Flow":[149],"fully":[152],"out-of-distribution":[153],"setting,":[154],"where":[155],"datasets":[160],"are":[161],"during":[163],"training.":[164],"Experimental":[165],"results":[166],"show":[167],"our":[169],"method":[170],"significantly":[171],"outperforms":[172],"SOTA":[173],"detectors.":[174],"Project":[175],"page:":[176],"https://opontorno.github.io/MuFlow.":[177]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-07-01T00:00:00"}