{"id":"https://openalex.org/W7138899716","doi":"https://doi.org/10.1609/aaai.v40i39.40607","title":"SafetyReminder: Reviving Delayed Safety Awareness of Vision-Language Models to Defend Against Jailbreak Attacks","display_name":"SafetyReminder: Reviving Delayed Safety Awareness of Vision-Language Models to Defend Against Jailbreak Attacks","publication_year":2026,"publication_date":"2026-03-14","ids":{"openalex":"https://openalex.org/W7138899716","doi":"https://doi.org/10.1609/aaai.v40i39.40607"},"language":null,"primary_location":{"id":"doi:10.1609/aaai.v40i39.40607","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i39.40607","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.1609/aaai.v40i39.40607","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5130057611","display_name":"Peiyuan Tang","orcid":null},"institutions":[{"id":"https://openalex.org/I87445476","display_name":"Xi'an Jiaotong University","ror":"https://ror.org/017zhmm22","country_code":"CN","type":"education","lineage":["https://openalex.org/I87445476"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Peiyuan Tang","raw_affiliation_strings":["Xi'an Jiaotong University"],"affiliations":[{"raw_affiliation_string":"Xi'an Jiaotong University","institution_ids":["https://openalex.org/I87445476"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Haojie Xin","orcid":null},"institutions":[{"id":"https://openalex.org/I87445476","display_name":"Xi'an Jiaotong University","ror":"https://ror.org/017zhmm22","country_code":"CN","type":"education","lineage":["https://openalex.org/I87445476"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Haojie Xin","raw_affiliation_strings":["Xi'an Jiaotong University"],"affiliations":[{"raw_affiliation_string":"Xi'an Jiaotong University","institution_ids":["https://openalex.org/I87445476"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100430876","display_name":"Xi Zhang","orcid":"https://orcid.org/0000-0003-3415-5345"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaodong Zhang","raw_affiliation_strings":["University of Science and Technology of China"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130053572","display_name":"Jun Sun","orcid":null},"institutions":[{"id":"https://openalex.org/I79891267","display_name":"Singapore Management University","ror":"https://ror.org/050qmg959","country_code":"SG","type":"education","lineage":["https://openalex.org/I79891267"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Jun Sun","raw_affiliation_strings":["Singapore Management University"],"affiliations":[{"raw_affiliation_string":"Singapore Management University","institution_ids":["https://openalex.org/I79891267"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129859992","display_name":"Qin Xia","orcid":null},"institutions":[{"id":"https://openalex.org/I87445476","display_name":"Xi'an Jiaotong University","ror":"https://ror.org/017zhmm22","country_code":"CN","type":"education","lineage":["https://openalex.org/I87445476"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qin Xia","raw_affiliation_strings":["Xi'an Jiaotong University"],"affiliations":[{"raw_affiliation_string":"Xi'an Jiaotong University","institution_ids":["https://openalex.org/I87445476"]}]},{"author_position":"last","author":{"id":null,"display_name":"Zijiang James Yang","orcid":null},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zijiang James Yang","raw_affiliation_strings":["University of Science and Technology of China"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China","institution_ids":["https://openalex.org/I126520041"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5130057611"],"corresponding_institution_ids":["https://openalex.org/I87445476"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.74238642,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"40","issue":"39","first_page":"33223","last_page":"33231"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9359999895095825,"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":0.9359999895095825,"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/T10883","display_name":"Ethics and Social Impacts of AI","score":0.010300000198185444,"subfield":{"id":"https://openalex.org/subfields/3311","display_name":"Safety Research"},"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.008700000122189522,"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":0.7271999716758728},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.6353999972343445},{"id":"https://openalex.org/keywords/phenomenon","display_name":"Phenomenon","score":0.5920000076293945},{"id":"https://openalex.org/keywords/perception","display_name":"Perception","score":0.5640000104904175},{"id":"https://openalex.org/keywords/helpfulness","display_name":"Helpfulness","score":0.5414000153541565},{"id":"https://openalex.org/keywords/threat-model","display_name":"Threat model","score":0.40630000829696655}],"concepts":[{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.7271999716758728},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6495000123977661},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.6353999972343445},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.6162999868392944},{"id":"https://openalex.org/C50335755","wikidata":"https://www.wikidata.org/wiki/Q483247","display_name":"Phenomenon","level":2,"score":0.5920000076293945},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.5640000104904175},{"id":"https://openalex.org/C2781265381","wikidata":"https://www.wikidata.org/wiki/Q5710255","display_name":"Helpfulness","level":2,"score":0.5414000153541565},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.4602000117301941},{"id":"https://openalex.org/C108827166","wikidata":"https://www.wikidata.org/wiki/Q175975","display_name":"Internet privacy","level":1,"score":0.41440001130104065},{"id":"https://openalex.org/C140547941","wikidata":"https://www.wikidata.org/wiki/Q7797194","display_name":"Threat model","level":2,"score":0.40630000829696655},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.3675999939441681},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.34599998593330383},{"id":"https://openalex.org/C145804949","wikidata":"https://www.wikidata.org/wiki/Q478123","display_name":"Situation awareness","level":2,"score":0.29260000586509705},{"id":"https://openalex.org/C132835097","wikidata":"https://www.wikidata.org/wiki/Q7663745","display_name":"System safety","level":2,"score":0.2734000086784363},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.26420000195503235},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.2628999948501587},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.2549999952316284}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1609/aaai.v40i39.40607","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i39.40607","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1609/aaai.v40i39.40607","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i39.40607","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"sustainable_development_goals":[{"score":0.7335711121559143,"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Vision-Language":[0],"Models":[1,6],"(VLMs)":[2],"extend":[3],"Large":[4],"Language":[5],"(LLMs)":[7],"with":[8,178],"visual":[9,28],"perception":[10],"capabilities,":[11],"unlocking":[12],"broad":[13],"applications":[14],"across":[15,162],"many":[16],"domains.":[17],"However,":[18],"ensuring":[19],"their":[20],"safety":[21,49,81,133],"remains":[22],"a":[23,44,52,103,110,182],"critical":[24],"challenge,":[25],"as":[26],"adversarial":[27,166],"inputs":[29],"can":[30],"easily":[31],"bypass":[32],"built-in":[33],"safeguards":[34],"and":[35,161,184],"elicit":[36],"harmful":[37,57,143,159],"content.":[38],"In":[39],"this":[40,73,98],"paper,":[41],"we":[42,46,100],"uncover":[43],"phenomenon":[45,74],"call":[47],"delayed":[48],"awareness,":[50],"where":[51],"jailbroken":[53],"VLM":[54],"initially":[55],"produces":[56],"content":[58,144],"but":[59],"ultimately":[60],"recognizes":[61],"the":[62,65,68,76,79,90,127,132,136],"harmfulness":[63],"at":[64],"end":[66],"of":[67,93,135,165,190],"generation":[69],"process.":[70],"We":[71,152],"attribute":[72],"to":[75,89,130],"fact":[77],"that":[78,108,171],"model's":[80],"awareness":[82,134],"against":[83],"jailbreaks":[84],"cannot":[85],"be":[86],"effectively":[87],"transferred":[88],"intermediate":[91],"stages":[92],"text":[94,129],"generation.":[95],"Motivated":[96],"by":[97],"insight,":[99],"introduce":[101],"SafetyReminder,":[102],"simple":[104],"yet":[105],"effective":[106],"defense":[107,176],"optimizes":[109],"learnable":[111],"soft":[112,122],"prompt":[113,123],"using":[114],"our":[115,154,172],"proposed":[116],"Safety-Activation":[117],"Prompt":[118],"Tuning":[119],"(SAPT).":[120],"This":[121],"is":[124],"inserted":[125],"into":[126],"generated":[128],"activate":[131],"model,":[137],"steering":[138],"it":[139],"toward":[140],"refusal":[141],"when":[142],"arises":[145],"while":[146],"preserving":[147],"helpfulness":[148],"in":[149],"benign":[150],"scenarios.":[151],"evaluate":[153],"method":[155,173],"on":[156],"three":[157,163],"established":[158],"benchmarks":[160],"types":[164],"attacks.":[167],"Experimental":[168],"results":[169],"demonstrate":[170],"achieves":[174],"state-of-the-art":[175],"performance":[177],"strong":[179],"generalization,":[180],"offering":[181],"practical":[183],"lightweight":[185],"solution":[186],"for":[187],"safe":[188],"deployment":[189],"VLMs.":[191]},"counts_by_year":[],"updated_date":"2026-04-21T08:09:41.155169","created_date":"2026-03-20T00:00:00"}
