{"id":"https://openalex.org/W4416959442","doi":"https://doi.org/10.48550/arxiv.2512.01185","title":"DefenSee: Dissecting Threat from Sight and Text -- A Multi-View Defensive Pipeline for Multi-modal Jailbreaks","display_name":"DefenSee: Dissecting Threat from Sight and Text -- A Multi-View Defensive Pipeline for Multi-modal Jailbreaks","publication_year":2025,"publication_date":"2025-12-01","ids":{"openalex":"https://openalex.org/W4416959442","doi":"https://doi.org/10.48550/arxiv.2512.01185"},"language":null,"primary_location":{"id":"pmh:oai:arXiv.org:2512.01185","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.01185","pdf_url":"https://arxiv.org/pdf/2512.01185","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2512.01185","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100413108","display_name":"Zihao Wang","orcid":"https://orcid.org/0000-0003-4822-0486"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Zihao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059071970","display_name":"Kar Wai Fok","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fok, Kar Wai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5040321131","display_name":"Vrizlynn L. L. Thing","orcid":"https://orcid.org/0000-0003-4424-8596"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Thing, Vrizlynn L. L.","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":true,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.8632000088691711,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.8632000088691711,"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/T11241","display_name":"Advanced Malware Detection Techniques","score":0.02590000070631504,"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/T10028","display_name":"Topic Modeling","score":0.010599999688565731,"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/robustness","display_name":"Robustness (evolution)","score":0.7437000274658203},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.5873000025749207},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.5595999956130981},{"id":"https://openalex.org/keywords/consistency","display_name":"Consistency (knowledge bases)","score":0.42640000581741333},{"id":"https://openalex.org/keywords/sight","display_name":"Sight","score":0.4115999937057495},{"id":"https://openalex.org/keywords/transcription","display_name":"Transcription (linguistics)","score":0.3776000142097473}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.760699987411499},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.7437000274658203},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.5873000025749207},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.5595999956130981},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5228000283241272},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.42640000581741333},{"id":"https://openalex.org/C1517167","wikidata":"https://www.wikidata.org/wiki/Q1134322","display_name":"Sight","level":2,"score":0.4115999937057495},{"id":"https://openalex.org/C179926584","wikidata":"https://www.wikidata.org/wiki/Q207714","display_name":"Transcription (linguistics)","level":2,"score":0.3776000142097473},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.35920000076293945},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.31610000133514404},{"id":"https://openalex.org/C140547941","wikidata":"https://www.wikidata.org/wiki/Q7797194","display_name":"Threat model","level":2,"score":0.3151000142097473},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2883000075817108},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.287200003862381},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2815999984741211},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.2669000029563904},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2581999897956848},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.25040000677108765}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2512.01185","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.01185","pdf_url":"https://arxiv.org/pdf/2512.01185","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2512.01185","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2512.01185","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":"pmh:oai:arXiv.org:2512.01185","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.01185","pdf_url":"https://arxiv.org/pdf/2512.01185","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416959442.pdf","grobid_xml":"https://content.openalex.org/works/W4416959442.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Multi-modal":[0],"large":[1],"language":[2],"models":[3],"(MLLMs),":[4],"capable":[5],"of":[6,119],"processing":[7],"text,":[8,39],"images,":[9],"and":[10,25,71,82,94],"audio,":[11],"have":[12],"been":[13],"widely":[14],"adopted":[15],"in":[16],"various":[17],"AI":[18],"applications.":[19],"However,":[20],"recent":[21],"MLLMs":[22,50],"integrating":[23],"images":[24],"text":[26],"remain":[27],"highly":[28],"vulnerable":[29],"to":[30,54,112,122],"coordinated":[31],"jailbreaks.":[32],"Existing":[33],"defenses":[34],"primarily":[35],"focus":[36],"on":[37,90,108,125],"the":[38,117,128,136],"lacking":[40],"robust":[41,70],"multi-modal":[42,73,92],"protection.":[43],"As":[44],"a":[45,69],"result,":[46],"studies":[47],"indicate":[48],"that":[49,77,98],"are":[51],"more":[52],"susceptible":[53],"malicious":[55],"or":[56],"unsafe":[57],"instructions,":[58],"unlike":[59],"their":[60],"text-only":[61],"counterparts.":[62],"In":[63],"this":[64],"paper,":[65],"we":[66],"proposed":[67],"DefenSee,":[68],"lightweight":[72],"black-box":[74],"defense":[75],"technique":[76],"leverages":[78],"image":[79],"variants":[80],"transcription":[81],"cross-modal":[83],"consistency":[84],"checks,":[85],"mimicking":[86],"human":[87],"judgment.":[88],"Experiments":[89],"popular":[91],"jailbreak":[93,120],"benign":[95,109],"datasets":[96],"show":[97],"DefenSee":[99],"consistently":[100],"enhances":[101],"MLLM":[102],"robustness":[103],"while":[104],"better":[105],"preserving":[106],"performance":[107],"tasks":[110],"compared":[111],"SOTA":[113],"defenses.":[114],"It":[115],"reduces":[116],"ASR":[118],"attacks":[121],"below":[123],"1.70%":[124],"MiniGPT4":[126],"using":[127],"MM-SafetyBench":[129],"benchmark,":[130],"significantly":[131],"outperforming":[132],"prior":[133],"methods":[134],"under":[135],"same":[137],"conditions.":[138]},"counts_by_year":[],"updated_date":"2026-08-21T09:56:20.448147","created_date":"2025-12-03T00:00:00"}