{"id":"https://openalex.org/W7164896953","doi":"https://doi.org/10.48550/arxiv.2606.15880","title":"Deep Residual Injection for Full-Spectrum Forensic Signal Perception in Multimodal Large Language Models","display_name":"Deep Residual Injection for Full-Spectrum Forensic Signal Perception in Multimodal Large Language Models","publication_year":2026,"publication_date":"2026-06-14","ids":{"openalex":"https://openalex.org/W7164896953","doi":"https://doi.org/10.48550/arxiv.2606.15880"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.15880","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.15880","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.2606.15880","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138721426","display_name":"Kaiqing Lin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Kaiqing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138708499","display_name":"Zhiyuan Yan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yan, Zhiyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138710758","display_name":"Ruoxin Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Ruoxin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103277929","display_name":"Ke-Yue Zhang","orcid":"https://orcid.org/0000-0003-3589-5580"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Ke-Yue","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138696444","display_name":"Yue Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Yue","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138749618","display_name":"Caiyong Piao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Piao, Caiyong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138743321","display_name":"Bin Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Bin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138710352","display_name":"Taiping Yao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yao, Taiping","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138721345","display_name":"Bo Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Bo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133363812","display_name":"Y H Xiao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiao, Youchang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138740124","display_name":"Shouhong Ding","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ding, Shouhong","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/T12357","display_name":"Digital Media Forensic Detection","score":0.1251000016927719,"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/T12357","display_name":"Digital Media Forensic Detection","score":0.1251000016927719,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.12489999830722809,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.09849999845027924,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.656499981880188},{"id":"https://openalex.org/keywords/residual","display_name":"Residual","score":0.6129999756813049},{"id":"https://openalex.org/keywords/semantic-similarity","display_name":"Semantic similarity","score":0.41190001368522644},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.40700000524520874},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.40459999442100525},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.3889999985694885},{"id":"https://openalex.org/keywords/signal","display_name":"SIGNAL (programming language)","score":0.3758000135421753},{"id":"https://openalex.org/keywords/perception","display_name":"Perception","score":0.3714999854564667},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.36649999022483826}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7922000288963318},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.656499981880188},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6158000230789185},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.6129999756813049},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.44449999928474426},{"id":"https://openalex.org/C130318100","wikidata":"https://www.wikidata.org/wiki/Q2268914","display_name":"Semantic similarity","level":2,"score":0.41190001368522644},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.40700000524520874},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.40459999442100525},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4004000127315521},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.3889999985694885},{"id":"https://openalex.org/C2779843651","wikidata":"https://www.wikidata.org/wiki/Q7390335","display_name":"SIGNAL (programming language)","level":2,"score":0.3758000135421753},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.3714999854564667},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.36649999022483826},{"id":"https://openalex.org/C197914299","wikidata":"https://www.wikidata.org/wiki/Q18650","display_name":"Semantic memory","level":3,"score":0.3418000042438507},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3375999927520752},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3206999897956848},{"id":"https://openalex.org/C90312973","wikidata":"https://www.wikidata.org/wiki/Q7449052","display_name":"Semantic data model","level":2,"score":0.31790000200271606},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3176000118255615},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.3082999885082245},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.3050999939441681},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.30309998989105225},{"id":"https://openalex.org/C178253425","wikidata":"https://www.wikidata.org/wiki/Q162668","display_name":"Visual perception","level":3,"score":0.28610000014305115},{"id":"https://openalex.org/C2779010991","wikidata":"https://www.wikidata.org/wiki/Q2720909","display_name":"Artifact (error)","level":2,"score":0.2833999991416931},{"id":"https://openalex.org/C511149849","wikidata":"https://www.wikidata.org/wiki/Q7449051","display_name":"Semantic computing","level":3,"score":0.2802000045776367},{"id":"https://openalex.org/C104267543","wikidata":"https://www.wikidata.org/wiki/Q208163","display_name":"Signal processing","level":3,"score":0.26109999418258667},{"id":"https://openalex.org/C2780992000","wikidata":"https://www.wikidata.org/wiki/Q17016113","display_name":"Generator (circuit theory)","level":3,"score":0.2596000134944916},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.25949999690055847},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.2587999999523163}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.15880","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.15880","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.2606.15880","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.15880","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":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education","score":0.6751317381858826}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Multimodal":[0],"large":[1],"language":[2],"models":[3],"(MLLMs)":[4],"have":[5],"been":[6],"increasingly":[7],"adopted":[8],"in":[9,63,72],"forensics":[10],"for":[11,27,79],"their":[12],"robust":[13,160],"semantic":[14,51,67,84,100,120,136],"understanding.":[15],"As":[16],"AI-generated":[17],"images":[18],"become":[19],"realistic,":[20],"semantic-level":[21],"inconsistencies":[22],"alone":[23],"are":[24,117,180],"often":[25],"insufficient":[26],"reliable":[28],"detection.":[29],"This":[30,129],"motivates":[31],"a":[32,56,108],"critical":[33],"question:":[34],"whether":[35],"MLLMs":[36],"can":[37],"achieve":[38],"full-spectrum":[39],"forensic":[40,60,140,153],"signal":[41,61],"perception,":[42],"i.e.,":[43],"capturing":[44],"low-level":[45],"generator":[46],"artifacts":[47],"without":[48],"sacrificing":[49],"pre-trained":[50],"knowledge.":[52],"We":[53],"further":[54],"perform":[55],"layer-wise":[57],"analysis":[58],"of":[59,152],"perception":[62],"MLLMs,":[64],"showing":[65],"that":[66,168],"information":[68],"is":[69],"primarily":[70],"formed":[71],"the":[73,144,157],"early-to-middle":[74],"layers,":[75],"whereas":[76],"direct":[77],"fine-tuning":[78],"artifact":[80],"learning":[81],"disrupts":[82],"these":[83],"representations.":[85],"Based":[86],"on":[87,156],"this":[88],"insight,":[89],"we":[90],"propose":[91],"Deep":[92],"Visual":[93],"Residual":[94],"MLLM":[95],"(Deep-VRM)":[96],"to":[97,133,147],"preserve":[98],"early":[99],"processing":[101],"while":[102],"injecting":[103],"artifact-specific":[104],"visual":[105],"signals":[106,154],"as":[107],"residual":[109],"path":[110],"into":[111],"an":[112],"intermediate":[113],"layer,":[114],"where":[115],"they":[116],"fused":[118],"with":[119],"token":[121],"representations":[122],"and":[123,138,142,161,178],"propagated":[124],"through":[125],"subsequent":[126],"trainable":[127],"layers.":[128],"enables":[130],"later":[131],"layers":[132],"jointly":[134],"model":[135,145],"reasoning":[137],"signal-level":[139],"cues,":[141],"surprisingly,":[143],"learns":[146],"adaptively":[148],"leverage":[149],"different":[150],"levels":[151],"depending":[155],"input,":[158],"achieving":[159],"generalizable":[162],"detection":[163],"performance.":[164],"Extensive":[165],"experiments":[166],"show":[167],"our":[169],"method":[170],"achieves":[171],"state-of-the-art":[172],"across":[173],"most":[174],"benchmarks.":[175],"The":[176],"code":[177],"data":[179],"available":[181],"at":[182],"https://github.com/KQL11/Deep-VRM.":[183]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-17T00:00:00"}