{"id":"https://openalex.org/W4416773970","doi":"https://doi.org/10.14428/esann/2026.es2026-153","title":"Adversarial Confusion Attack: Disrupting Multimodal Large Language Models","display_name":"Adversarial Confusion Attack: Disrupting Multimodal Large Language Models","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W4416773970","doi":"https://doi.org/10.14428/esann/2026.es2026-153"},"language":null,"primary_location":{"id":"doi:10.14428/esann/2026.es2026-153","is_oa":true,"landing_page_url":"https://doi.org/10.14428/esann/2026.es2026-153","pdf_url":"https://doi.org/10.14428/esann/2026.es2026-153","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ESANN 2026 proceesdings","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.14428/esann/2026.es2026-153","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5003640619","display_name":"Jakub Ho\u015bci\u0142owicz","orcid":null},"institutions":[{"id":"https://openalex.org/I108403487","display_name":"Warsaw University of Technology","ror":"https://ror.org/00y0xnp53","country_code":"PL","type":"education","lineage":["https://openalex.org/I108403487"]},{"id":"https://openalex.org/I4210125529","display_name":"National Institute of Telecommunications","ror":"https://ror.org/03053v606","country_code":"PL","type":"facility","lineage":["https://openalex.org/I4210125529"]}],"countries":["PL"],"is_corresponding":false,"raw_author_name":"Jakub Hoscilowicz","raw_affiliation_strings":["Warsaw University of Technology Institute of Telecommunications and Cybersecurity Nowowiejska 15/19 , Warsaw , Poland"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Warsaw University of Technology Institute of Telecommunications and Cybersecurity Nowowiejska 15/19 , Warsaw , Poland","institution_ids":["https://openalex.org/I108403487","https://openalex.org/I4210125529"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5084366475","display_name":"Artur Janicki","orcid":"https://orcid.org/0000-0002-9937-4402"},"institutions":[{"id":"https://openalex.org/I108403487","display_name":"Warsaw University of Technology","ror":"https://ror.org/00y0xnp53","country_code":"PL","type":"education","lineage":["https://openalex.org/I108403487"]},{"id":"https://openalex.org/I4210125529","display_name":"National Institute of Telecommunications","ror":"https://ror.org/03053v606","country_code":"PL","type":"facility","lineage":["https://openalex.org/I4210125529"]}],"countries":["PL"],"is_corresponding":false,"raw_author_name":"Artur Janicki","raw_affiliation_strings":["Warsaw University of Technology Institute of Telecommunications and Cybersecurity Nowowiejska 15/19 , Warsaw , Poland"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Warsaw University of Technology Institute of Telecommunications and Cybersecurity Nowowiejska 15/19 , Warsaw , Poland","institution_ids":["https://openalex.org/I108403487","https://openalex.org/I4210125529"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.02907622,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"345","last_page":"350"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9591000080108643,"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":0.9591000080108643,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.006500000134110451,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.005400000140070915,"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/adversarial-system","display_name":"Adversarial system","score":0.9621999859809875},{"id":"https://openalex.org/keywords/confusion","display_name":"Confusion","score":0.7121000289916992},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.5408999919891357},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5297999978065491},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.5113999843597412},{"id":"https://openalex.org/keywords/entropy","display_name":"Entropy (arrow of time)","score":0.36090001463890076}],"concepts":[{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.9621999859809875},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7121000289916992},{"id":"https://openalex.org/C2781140086","wikidata":"https://www.wikidata.org/wiki/Q557945","display_name":"Confusion","level":2,"score":0.7121000289916992},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.5408999919891357},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5297999978065491},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5293999910354614},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.5113999843597412},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.36090001463890076},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3257000148296356},{"id":"https://openalex.org/C9679016","wikidata":"https://www.wikidata.org/wiki/Q1417473","display_name":"Principle of maximum entropy","level":2,"score":0.31200000643730164},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2921000123023987},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.28839999437332153},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.27950000762939453},{"id":"https://openalex.org/C163339463","wikidata":"https://www.wikidata.org/wiki/Q484598","display_name":"CAPTCHA","level":2,"score":0.2556999921798706}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.14428/esann/2026.es2026-153","is_oa":true,"landing_page_url":"https://doi.org/10.14428/esann/2026.es2026-153","pdf_url":"https://doi.org/10.14428/esann/2026.es2026-153","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ESANN 2026 proceesdings","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2511.20494","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2511.20494","pdf_url":"https://arxiv.org/pdf/2511.20494","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.2511.20494","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2511.20494","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":"doi:10.14428/esann/2026.es2026-153","is_oa":true,"landing_page_url":"https://doi.org/10.14428/esann/2026.es2026-153","pdf_url":"https://doi.org/10.14428/esann/2026.es2026-153","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ESANN 2026 proceesdings","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416773970.pdf","grobid_xml":"https://content.openalex.org/works/W4416773970.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0,121],"introduce":[1,122],"the":[2,22,31,67,82,86,105,123,145,158,162],"adversarial":[3,43,75,90,97,124],"confusion":[4,125],"attack":[5,56,106],"as":[6,100],"a":[7,61,73,95,151],"new":[8],"class":[9],"of":[10,64,140],"threats":[11],"against":[12],"multimodal":[13],"large":[14],"language":[15],"models":[16,80,156],"(MLLMs).Unlike":[17],"jailbreaks":[18],"or":[19,35],"targeted":[20],"misclassification,":[21],"goal":[23],"is":[24],"to":[25,47,111,131],"induce":[26],"systematic":[27],"disruption":[28],"that":[29,72,109,150],"makes":[30],"model":[32,142],"generate":[33],"incoherent":[34],"confidently":[36],"incorrect":[37],"outputs.Practical":[38],"applications":[39],"include":[40],"embedding":[41],"such":[42,99],"images":[44],"into":[45],"websites":[46],"prevent":[48],"MLLM-powered":[49],"AI":[50],"Agents":[51],"from":[52],"operating":[53],"reliably.The":[54],"proposed":[55],"maximizes":[57,128],"next-token":[58],"entropy":[59,130],"using":[60],"small":[62],"ensemble":[63],"open-source":[65,114],"MLLMs.In":[66],"white-box":[68,146],"setting,":[69,147],"we":[70,135,148],"show":[71,149],"single":[74,152],"image":[76],"can":[77],"disrupt":[78],"all":[79,155],"in":[81,85,157,160],"ensemble,":[83,159],"both":[84,112,161],"full-image":[87,163],"and":[88,117,134,164],"CAPTCHA-style":[89],"patch":[91],"settings.Despite":[92],"relying":[93],"on":[94],"basic":[96],"technique,":[98],"projected":[101],"gradient":[102],"descent":[103],"(PGD),":[104],"generates":[107],"perturbations":[108],"transfer":[110],"unseen":[113],"(e.g.,":[115,119],"Qwen3-VL)":[116],"proprietary":[118],"GPT-5.1)models.":[120],"attack,":[126],"which":[127],"output":[129],"destabilize":[132],"decoding,":[133],"characterize":[136],"five":[137],"distinct":[138],"modes":[139],"resulting":[141],"failure.":[143],"In":[144],"perturbation":[153],"disrupts":[154],"Adversarial":[165],"CAPTCHA":[166],"settings.":[167]},"counts_by_year":[],"updated_date":"2026-08-05T07:39:15.569665","created_date":"2025-11-28T00:00:00"}