{"id":"https://openalex.org/W7160961689","doi":"https://doi.org/10.48550/arxiv.2605.10334","title":"The Alpha Blending Hypothesis: Compositing Shortcut in Deepfake Detection","display_name":"The Alpha Blending Hypothesis: Compositing Shortcut in Deepfake Detection","publication_year":2026,"publication_date":"2026-05-11","ids":{"openalex":"https://openalex.org/W7160961689","doi":"https://doi.org/10.48550/arxiv.2605.10334"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.10334","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.10334","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.2605.10334","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5033869972","display_name":"Andrii Yermakov","orcid":"https://orcid.org/0000-0002-2173-5095"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yermakov, Andrii","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135969266","display_name":"Jan Cech","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cech, Jan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003887059","display_name":"Mario Fritz","orcid":"https://orcid.org/0000-0001-8949-9896"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fritz, Mario","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5135957238","display_name":"Jiri Matas","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Matas, Jiri","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":0.7383999824523926,"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":0.7383999824523926,"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":0.1941000074148178,"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/T11094","display_name":"Face Recognition and Perception","score":0.011500000022351742,"subfield":{"id":"https://openalex.org/subfields/2805","display_name":"Cognitive Neuroscience"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.71670001745224},{"id":"https://openalex.org/keywords/compositing","display_name":"Compositing","score":0.6960999965667725},{"id":"https://openalex.org/keywords/function","display_name":"Function (biology)","score":0.5012000203132629},{"id":"https://openalex.org/keywords/alpha","display_name":"Alpha (finance)","score":0.46309998631477356},{"id":"https://openalex.org/keywords/sensitivity","display_name":"Sensitivity (control systems)","score":0.45080000162124634},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.43309998512268066},{"id":"https://openalex.org/keywords/detector","display_name":"Detector","score":0.4027000069618225},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.38940000534057617}],"concepts":[{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.71670001745224},{"id":"https://openalex.org/C129315195","wikidata":"https://www.wikidata.org/wiki/Q1121886","display_name":"Compositing","level":3,"score":0.6960999965667725},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6926000118255615},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5580999851226807},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.5012000203132629},{"id":"https://openalex.org/C64943373","wikidata":"https://www.wikidata.org/wiki/Q2651003","display_name":"Alpha (finance)","level":4,"score":0.46309998631477356},{"id":"https://openalex.org/C21200559","wikidata":"https://www.wikidata.org/wiki/Q7451068","display_name":"Sensitivity (control systems)","level":2,"score":0.45080000162124634},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.43309998512268066},{"id":"https://openalex.org/C94915269","wikidata":"https://www.wikidata.org/wiki/Q1834857","display_name":"Detector","level":2,"score":0.4027000069618225},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.38940000534057617},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.37450000643730164},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3587000072002411},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.35370001196861267},{"id":"https://openalex.org/C64869954","wikidata":"https://www.wikidata.org/wiki/Q1859747","display_name":"False positive paradox","level":2,"score":0.34769999980926514},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.34279999136924744},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.326200008392334},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.29600000381469727},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.28119999170303345},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.26600000262260437},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.25099998712539673}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.10334","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.10334","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.2605.10334","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.10334","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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recent":[0],"deepfake":[1,63,106],"detection":[2],"methods":[3],"demonstrate":[4],"improved":[5],"cross-dataset":[6,101],"generalization,":[7],"yet":[8],"the":[9,16,48,59,69,79,98,153],"underlying":[10],"mechanisms":[11],"remain":[12],"underexplored.":[13],"We":[14,56,77],"introduce":[15],"Alpha":[17],"Blending":[18],"Hypothesis,":[19],"positing":[20],"that":[21,62,82,123],"state-of-the-art":[22,140],"frame-based":[23],"detectors":[24,64],"primarily":[25],"function":[26],"as":[27],"alpha":[28],"blending":[29,127,133],"searchers;":[30],"rather":[31],"than":[32],"learning":[33],"semantic":[34],"anomalies":[35],"or":[36],"specific":[37],"generative":[38],"neural":[39],"fingerprints,":[40],"they":[41],"localize":[42],"low-level":[43],"compositing":[44],"artifacts":[45],"introduced":[46],"during":[47,118],"integration":[49],"of":[50,88,142],"manipulated":[51],"faces":[52],"into":[53],"target":[54],"frames.":[55],"experimentally":[57],"validate":[58],"hypothesis,":[60],"demonstrating":[61],"exhibit":[65],"high":[66],"sensitivity":[67],"to":[68,132],"so-called":[70],"self-blended":[71],"images":[72,91],"(SBI)":[73],"and":[74,111,129,152],"non-generative":[75],"manipulations.":[76],"propose":[78],"method":[80],"BlenD":[81],"leverages":[83],"a":[84,139],"large-scale,":[85],"diverse":[86],"dataset":[87],"real-only":[89],"facial":[90],"augmented":[92],"with":[93,150],"SBI.":[94],"This":[95],"approach":[96],"achieves":[97],"best":[99],"average":[100],"generalization":[102],"on":[103],"15":[104],"compositional":[105],"datasets":[107],"released":[108],"between":[109],"2019":[110],"2025":[112],"without":[113],"utilizing":[114],"explicitly":[115],"generated":[116],"deepfakes":[117],"training.":[119],"Furthermore,":[120],"we":[121],"show":[122],"predictions":[124],"from":[125],"explicit":[126],"searchers":[128],"models":[130],"resilient":[131],"shortcuts":[134],"are":[135],"highly":[136],"complementary,":[137],"yielding":[138],"AUROC":[141],"94.0%":[143],"in":[144],"an":[145],"ensemble":[146],"configuration.":[147],"The":[148],"code":[149],"experiments":[151],"trained":[154],"model":[155],"will":[156],"be":[157],"publicly":[158],"released.":[159]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-13T00:00:00"}