{"id":"https://openalex.org/W7127902446","doi":"https://doi.org/10.48550/arxiv.2602.04702","title":"Fine-Grained Frame Modeling in Multi-head Self-Attention for Speech Deepfake Detection","display_name":"Fine-Grained Frame Modeling in Multi-head Self-Attention for Speech Deepfake Detection","publication_year":2026,"publication_date":"2026-02-04","ids":{"openalex":"https://openalex.org/W7127902446","doi":"https://doi.org/10.48550/arxiv.2602.04702"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2602.04702","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":null,"any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5125144106","display_name":"Tuan Dat Phuong","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Phuong, Tuan Dat","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091500888","display_name":"Duc-Tuan Truong","orcid":"https://orcid.org/0009-0002-1767-7598"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Truong, Duc-Tuan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5010668364","display_name":"Long-Vu Hoang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hoang, Long-Vu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"Thu, Trang Nguyen Thi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Thu, Trang Nguyen Thi","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5125144106"],"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.3578000068664551,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.3578000068664551,"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/T10860","display_name":"Speech and Audio Processing","score":0.2703999876976013,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.07989999651908875,"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/frame","display_name":"Frame (networking)","score":0.7149999737739563},{"id":"https://openalex.org/keywords/dimension","display_name":"Dimension (graph theory)","score":0.5232999920845032},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.33550000190734863},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3312999904155731},{"id":"https://openalex.org/keywords/baseline","display_name":"Baseline (sea)","score":0.328000009059906}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7936999797821045},{"id":"https://openalex.org/C126042441","wikidata":"https://www.wikidata.org/wiki/Q1324888","display_name":"Frame (networking)","level":2,"score":0.7149999737739563},{"id":"https://openalex.org/C33676613","wikidata":"https://www.wikidata.org/wiki/Q13415176","display_name":"Dimension (graph theory)","level":2,"score":0.5232999920845032},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5232999920845032},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3490000069141388},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3440999984741211},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.33550000190734863},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3312999904155731},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.328000009059906},{"id":"https://openalex.org/C167900197","wikidata":"https://www.wikidata.org/wiki/Q11081100","display_name":"Spoofing attack","level":2,"score":0.3206000030040741},{"id":"https://openalex.org/C520049643","wikidata":"https://www.wikidata.org/wiki/Q189760","display_name":"Voting","level":3,"score":0.2849000096321106},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.25360000133514404}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2602.04702","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"doi:10.48550/arxiv.2602.04702","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.04702","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":"article"}],"best_oa_location":{"id":"pmh:doi:10.48550/arxiv.2602.04702","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"sustainable_development_goals":[{"score":0.52894127368927,"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Transformer-based":[0],"models":[1],"have":[2],"shown":[3],"strong":[4],"performance":[5],"in":[6,35],"speech":[7,68,151],"deepfake":[8,31,69,152],"detection,":[9,70],"largely":[10],"due":[11],"to":[12,97,102],"the":[13,16,40,72,99,114,129,143],"effectiveness":[14,144],"of":[15,43,123,145],"multi-head":[17,81],"self-attention":[18],"(MHSA)":[19],"mechanism.":[20],"MHSA":[21],"provides":[22],"frame-level":[23],"attention":[24],"scores,":[25],"which":[26],"are":[27,76,88],"particularly":[28],"valuable":[29],"because":[30],"artifacts":[32],"often":[33],"occur":[34],"small,":[36],"localized":[37],"regions":[38],"along":[39],"temporal":[41],"dimension":[42],"speech.":[44],"This":[45],"makes":[46],"fine-grained":[47,62,147],"frame":[48,63],"modeling":[49,64,148],"essential":[50],"for":[51,66,149],"accurately":[52],"detecting":[53],"subtle":[54,104],"spoofing":[55,105],"cues.":[56,106],"In":[57],"this":[58],"work,":[59],"we":[60],"propose":[61],"(FGFM)":[65],"MHSA-based":[67],"where":[71],"most":[73],"informative":[74],"frames":[75,87],"first":[77],"selected":[78,86],"through":[79],"a":[80,92],"voting":[82],"(MHV)":[83],"module.":[84],"These":[85,136],"then":[89],"refined":[90],"via":[91],"cross-layer":[93],"refinement":[94],"(CLR)":[95],"module":[96],"enhance":[98],"model's":[100],"ability":[101],"learn":[103],"Experimental":[107],"results":[108],"demonstrate":[109],"that":[110],"our":[111,146],"method":[112],"outperforms":[113],"baseline":[115],"model":[116],"and":[117,126,132],"achieves":[118],"Equal":[119],"Error":[120],"Rate":[121],"(EER)":[122],"0.90%,":[124],"1.88%,":[125],"6.64%":[127],"on":[128],"LA21,":[130],"DF21,":[131],"ITW":[133],"datasets,":[134],"respectively.":[135],"consistent":[137],"improvements":[138],"across":[139],"multiple":[140],"benchmarks":[141],"highlight":[142],"robust":[150],"detection.":[153]},"counts_by_year":[],"updated_date":"2026-04-16T08:26:57.006410","created_date":"2026-02-07T00:00:00"}