{"id":"https://openalex.org/W7204139990","doi":"https://doi.org/10.18653/2026.findings-emnlp.348","title":"Vis-Poison: Poisoning Visual Knowledge in Multimodal Retrieval-Augmented Generation","display_name":"Vis-Poison: Poisoning Visual Knowledge in Multimodal Retrieval-Augmented Generation","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7204139990","doi":"https://doi.org/10.18653/2026.findings-emnlp.348"},"language":"en","primary_location":{"id":"doi:10.18653/2026.findings-emnlp.348","doi":"https://doi.org/10.18653/2026.findings-emnlp.348","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.findings-emnlp.348","pdf_url":"https://aclanthology.org/2026.findings-emnlp.348.pdf","source":{"id":"https://openalex.org/S7407096085","display_name":"Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2026","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","is_traditional":true,"indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"bronze","oa_url":"https://aclanthology.org/2026.findings-emnlp.348.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5148034001","display_name":"Rujin Liang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I204831749","display_name":"Southwestern University of Finance and Economics","ror":"https://ror.org/04ewct822","country_code":"CN","type":"education","lineage":["https://openalex.org/I204831749"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Rujin Liang","raw_affiliation_strings":["Southwestern University of Finance and Economics , Chengdu , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Southwestern University of Finance and Economics , Chengdu , China","institution_ids":["https://openalex.org/I204831749"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5147990910","display_name":"Zhongpu Chen","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I204831749","display_name":"Southwestern University of Finance and Economics","ror":"https://ror.org/04ewct822","country_code":"CN","type":"education","lineage":["https://openalex.org/I204831749"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhongpu Chen","raw_affiliation_strings":["Southwestern University of Finance and Economics , Chengdu , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Southwestern University of Finance and Economics , Chengdu , China","institution_ids":["https://openalex.org/I204831749"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5147986728","display_name":"Yuhao Lei","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I204831749","display_name":"Southwestern University of Finance and Economics","ror":"https://ror.org/04ewct822","country_code":"CN","type":"education","lineage":["https://openalex.org/I204831749"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuhao Lei","raw_affiliation_strings":["Southwestern University of Finance and Economics , Chengdu , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Southwestern University of Finance and Economics , Chengdu , China","institution_ids":["https://openalex.org/I204831749"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5123966830","display_name":"Xin Miao","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I36399199","display_name":"Nanjing University of Science and Technology","ror":"https://ror.org/00xp9wg62","country_code":"CN","type":"education","lineage":["https://openalex.org/I36399199"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xin Miao","raw_affiliation_strings":["Nanjing University of Science and Technology , Nanjing , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanjing University of Science and Technology , Nanjing , China","institution_ids":["https://openalex.org/I36399199"]}]}],"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":0.5546261,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"7280","last_page":"7294"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9990000128746033,"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":0.9990000128746033,"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/T12357","display_name":"Digital Media Forensic Detection","score":0.00039999998989515007,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":9.999999747378752e-05,"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/multimodal-retrieval-augmented-generation","display_name":"multimodal retrieval-augmented generation","score":0.9599999785423279},{"id":"https://openalex.org/keywords/multimodal-large-language-models","display_name":"multimodal large language models","score":0.9769999980926514},{"id":"https://openalex.org/keywords/adversarial-attacks","display_name":"adversarial attacks","score":0.7110000252723694},{"id":"https://openalex.org/keywords/black-box-attacks","display_name":"black-box attacks","score":0.7799999713897705},{"id":"https://openalex.org/keywords/image-captioning","display_name":"image captioning","score":0.5009999871253967}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7006999850273132},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5103999972343445},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.49459999799728394},{"id":"https://openalex.org/C117251300","wikidata":"https://www.wikidata.org/wiki/Q1849855","display_name":"Parametric statistics","level":2,"score":0.3619999885559082},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.32749998569488525},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3122999966144562},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3005000054836273},{"id":"https://openalex.org/C3017944768","wikidata":"https://www.wikidata.org/wiki/Q1450463","display_name":"Poison control","level":2,"score":0.28700000047683716},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2800000011920929},{"id":"https://openalex.org/C4441509","wikidata":"https://www.wikidata.org/wiki/Q6418787","display_name":"Multimodal therapy","level":2,"score":0.27730000019073486},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2705000042915344},{"id":"https://openalex.org/C89611455","wikidata":"https://www.wikidata.org/wiki/Q6804646","display_name":"Mechanism (biology)","level":2,"score":0.26829999685287476},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.25270000100135803}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/2026.findings-emnlp.348","doi":"https://doi.org/10.18653/2026.findings-emnlp.348","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.findings-emnlp.348","pdf_url":"https://aclanthology.org/2026.findings-emnlp.348.pdf","source":{"id":"https://openalex.org/S7407096085","display_name":"Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2026","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2608.20756","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.20756","pdf_url":"https://arxiv.org/pdf/2608.20756","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,"listed_in":[],"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.2608.20756","doi":"https://doi.org/10.48550/arxiv.2608.20756","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.20756","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,"listed_in":[],"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"}],"best_oa_location":{"id":"doi:10.18653/2026.findings-emnlp.348","doi":"https://doi.org/10.18653/2026.findings-emnlp.348","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.findings-emnlp.348","pdf_url":"https://aclanthology.org/2026.findings-emnlp.348.pdf","source":{"id":"https://openalex.org/S7407096085","display_name":"Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204139990.pdf","grobid_xml":"https://content.openalex.org/works/W7204139990.grobid-xml"},"referenced_works_count":7,"referenced_works":["https://openalex.org/W4404341467","https://openalex.org/W4414587373","https://openalex.org/W4414855245","https://openalex.org/W4416780398","https://openalex.org/W4417511252","https://openalex.org/W7133355669","https://openalex.org/W7139046133"],"related_works":["https://openalex.org/W6910602221","https://openalex.org/W6910642639","https://openalex.org/W7133355669","https://openalex.org/W6967388679","https://openalex.org/W4414855245","https://openalex.org/W7161092171","https://openalex.org/W7152468372","https://openalex.org/W7106284304","https://openalex.org/W7213288484","https://openalex.org/W7135424608"],"abstract_inverted_index":{"While":[0],"multimodal":[1,23,88,111],"retrieval-augmented":[2],"generation":[3,96],"(RAG)":[4],"systems":[5],"increasingly":[6],"rely":[7,32],"on":[8,33,148],"images":[9],"as":[10],"external":[11],"knowledge":[12,43,112,129],"sources,":[13],"the":[14,47,52,167],"introduction":[15],"of":[16,105,169],"poisoned":[17,48,77],"visual":[18,42],"evidence":[19],"can":[20,124],"severely":[21],"compromise":[22],"large":[24],"language":[25],"model":[26],"(MLLM)":[27],"generation.Unlike":[28],"prior":[29],"attacks":[30],"that":[31,73,123],"altering":[34],"textual":[35],"metadata,":[36,59],"we":[37,82],"introduce":[38],"Vis-Poison,":[39],"a":[40,149,161,170],"novel":[41],"poisoning":[44],"attack":[45,65,102],"where":[46],"image":[49],"itself":[50],"is":[51,66,179],"attacker-controlled":[53],"payload,":[54],"without":[55],"manipulating":[56],"captions,":[57],"summaries,":[58],"or":[60],"other":[61],"associated":[62],"text.Specifically,":[63],"this":[64],"instantiated":[67],"through":[68],"an":[69,100,132],"automated":[70],"multi-agent":[71],"method":[72],"constructs":[74],"visually":[75],"plausible":[76],"images.To":[78],"assess":[79],"its":[80,152,177],"impact,":[81],"evaluate":[83],"Vis-Poison":[84,98,117],"across":[85],"two":[86],"representative":[87],"RAG":[89],"pipelines,":[90],"four":[91],"embedding":[92],"models,":[93],"and":[94,138,156],"six":[95],"models.Empirically,":[97],"achieves":[99],"end-to-end":[101],"success":[103,134],"rate":[104,135],"40.16%":[106],"to":[107],"65.40%":[108],"against":[109,120,160],"30k-entry":[110],"bases":[113],"in":[114],"black-box":[115],"settings.Moreover,":[116],"remains":[118],"effective":[119],"various":[121],"MLLMs":[122],"answer":[125],"correctly":[126],"from":[127],"parametric":[128],"alone,":[130],"with":[131],"average":[133],"above":[136],"60%.Code":[137],"data":[139],"are":[140],"available":[141],"at":[142],"https://github.":[143],"com/SWUFE-DB-Group/Vis-Poison.A":[144],"blue-chinned":[145,172],"sapphire":[146,173],"perches":[147],"thin":[150],"branch,":[151],"shimmering":[153],"emeraldblue":[154],"feathers":[155],"iridescent":[157],"bill":[158],"glowing":[159],"soft":[162],"green":[163],"background.Clean":[164],"Poisoned":[165],"Is":[166],"beak":[168],"male":[171],"bird":[174],"thicker":[175],"than":[176],"eye":[178],"wide?Correct":[180],"Answer:":[181,184],"No":[182],"Wrong":[183],"Yes":[185],"...":[186]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-08-25T00:00:00"}