{"id":"https://openalex.org/W7154221166","doi":"https://doi.org/10.48550/arxiv.2604.09711","title":"Head-wise Modality Specialization within MLLMs for Robust Fake News Detection under Missing Modality","display_name":"Head-wise Modality Specialization within MLLMs for Robust Fake News Detection under Missing Modality","publication_year":2026,"publication_date":"2026-04-08","ids":{"openalex":"https://openalex.org/W7154221166","doi":"https://doi.org/10.48550/arxiv.2604.09711"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.09711","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.09711","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":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.2604.09711","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5133607360","display_name":"Kai Qian","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qian, Kai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133595967","display_name":"Weijie Shi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shi, Weijie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133595833","display_name":"Jiaqi Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Jiaqi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101737334","display_name":"Mengze Li","orcid":"https://orcid.org/0000-0002-3482-234X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Mengze","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133580156","display_name":"Hao Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Hao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129679091","display_name":"Yue Cui","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cui, Yue","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103010744","display_name":"Hanghui Guo","orcid":"https://orcid.org/0009-0002-2659-1550"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guo, Hanghui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133562576","display_name":"Ziyi Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Ziyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133580879","display_name":"Jia Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Jia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100940170","display_name":"Jiajie Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Jiajie","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/T11147","display_name":"Misinformation and Its Impacts","score":0.9922000169754028,"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"}},"topics":[{"id":"https://openalex.org/T11147","display_name":"Misinformation and Its Impacts","score":0.9922000169754028,"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/T14347","display_name":"Big Data and Digital Economy","score":0.0008999999845400453,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T11644","display_name":"Spam and Phishing Detection","score":0.0007999999797903001,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/modality","display_name":"Modality (human\u2013computer interaction)","score":0.8264999985694885},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.7409999966621399},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.6671000123023987},{"id":"https://openalex.org/keywords/credibility","display_name":"Credibility","score":0.6586999893188477},{"id":"https://openalex.org/keywords/modalities","display_name":"Modalities","score":0.6327000260353088},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5144000053405762}],"concepts":[{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.8264999985694885},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7795000076293945},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.7409999966621399},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6671000123023987},{"id":"https://openalex.org/C2780224610","wikidata":"https://www.wikidata.org/wiki/Q1530061","display_name":"Credibility","level":2,"score":0.6586999893188477},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6471999883651733},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.6327000260353088},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5144000053405762},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.4171000123023987},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3837999999523163},{"id":"https://openalex.org/C9357733","wikidata":"https://www.wikidata.org/wiki/Q6878417","display_name":"Missing data","level":2,"score":0.35830000042915344},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.27649998664855957},{"id":"https://openalex.org/C2780660688","wikidata":"https://www.wikidata.org/wiki/Q25052564","display_name":"Multimodal learning","level":2,"score":0.263700008392334},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2572000026702881}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.09711","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.09711","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":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.2604.09711","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.09711","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Multimodal":[0,76],"fake":[1],"news":[2,8,19],"detection":[3,37],"(MFND)":[4],"aims":[5],"to":[6,27,55,124,144,156],"verify":[7],"credibility":[9],"by":[10],"jointly":[11],"exploiting":[12],"textual":[13],"and":[14,32,62,96,147],"visual":[15],"evidence.":[16],"However,":[17],"real-world":[18],"dissemination":[20],"frequently":[21],"suffers":[22],"from":[23,173,176,181],"missing":[24,85,102,192],"modality":[25,61,118,193],"due":[26,54],"deleted":[28],"images,":[29],"corrupted":[30],"screenshots,":[31],"similar":[33],"issues.":[34],"Thus,":[35],"robust":[36,82],"in":[38,52,94],"this":[39,68,122],"scenario":[40],"requires":[41],"preserving":[42,195],"strong":[43],"verification":[44,114,127],"ability":[45,115,128],"for":[46,81,129],"each":[47],"modality,":[48,103,132],"which":[49],"is":[50],"challenging":[51],"MFND":[53,83],"insufficient":[56],"learning":[57],"of":[58,112],"the":[59,130,177],"low-contribution":[60,131],"scarce":[63,159],"unimodal":[64,113,160,178],"annotations.":[65],"To":[66],"address":[67],"issue,":[69],"we":[70,88,133,162],"propose":[71,163],"Head-wise":[72],"Modality":[73],"Specialization":[74],"within":[75],"Large":[77],"Language":[78],"Models":[79],"(MLLMs)":[80],"under":[84,101,191],"modality.":[86],"Specifically,":[87],"first":[89],"systematically":[90],"study":[91],"attention":[92,153],"heads":[93,107,143,172],"MLLMs":[95],"their":[97,117,149],"relationship":[98],"with":[99,197],"performance":[100,196],"showing":[104],"that":[105,139,169,186],"modality-critical":[106],"serve":[108],"as":[109],"key":[110],"carriers":[111],"through":[116,151],"specialization.":[119],"Based":[120],"on":[121],"observation,":[123],"better":[125,157],"preserve":[126],"introduce":[134],"a":[135,164],"head-wise":[136],"specialization":[137,150],"mechanism":[138],"explicitly":[140],"allocates":[141],"these":[142,171],"different":[145],"modalities":[146],"preserves":[148],"lower-bound":[152],"constraints.":[154],"Furthermore,":[155],"exploit":[158],"annotations,":[161],"Unimodal":[165],"Knowledge":[166],"Retention":[167],"strategy":[168],"prevents":[170],"drifting":[174],"away":[175],"knowledge":[179],"learned":[180],"limited":[182],"supervision.":[183],"Experiments":[184],"show":[185],"our":[187],"method":[188],"improves":[189],"robustness":[190],"while":[194],"full":[198],"multimodal":[199],"input.":[200]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-15T00:00:00"}