{"id":"https://openalex.org/W4417423213","doi":"https://doi.org/10.1145/3856301","title":"From LLMs to MLLMs to Agents: A Survey of Emerging Paradigms in Jailbreak Attacks and Defenses within LLM Ecosystem","display_name":"From LLMs to MLLMs to Agents: A Survey of Emerging Paradigms in Jailbreak Attacks and Defenses within LLM Ecosystem","publication_year":2026,"publication_date":"2026-10-29","ids":{"openalex":"https://openalex.org/W4417423213","doi":"https://doi.org/10.1145/3856301"},"language":"en","primary_location":{"id":"doi:10.1145/3856301","doi":"https://doi.org/10.1145/3856301","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3856301","pdf_url":null,"source":{"id":"https://openalex.org/S41523882","display_name":"ACM Transactions on Knowledge Discovery from Data","issn_l":"1556-4681","issn":["1556-4681","1556-472X"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["ccf-b","cwts-core","dongbi-a","jufo-3","norway-1","poland-100","russia-white-list-1"],"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Knowledge Discovery from Data","raw_type":"journal-article"},"volume":null,"type":"article","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2506.15170","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Yanxu Mao","orcid":"https://orcid.org/0009-0009-7906-8561","observed_orcids":["https://orcid.org/0009-0009-7906-8561"]},"institutions":[{"id":"https://openalex.org/I173899330","display_name":"Henan University","ror":"https://ror.org/003xyzq10","country_code":"CN","type":"education","lineage":["https://openalex.org/I173899330"]},{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanxu Mao","raw_affiliation_strings":["School of Computer Science, Shanghai Jiao Tong University, Shanghai, China and School of Software, Henan University, Kaifeng, China"],"raw_orcid":"https://orcid.org/0009-0009-7906-8561","affiliations":[{"raw_affiliation_string":"School of Computer Science, Shanghai Jiao Tong University, Shanghai, China and School of Software, Henan University, Kaifeng, China","institution_ids":["https://openalex.org/I173899330","https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Tiehan Cui","orcid":"https://orcid.org/0009-0003-7429-8452","observed_orcids":["https://orcid.org/0009-0003-7429-8452"]},"institutions":[{"id":"https://openalex.org/I173899330","display_name":"Henan University","ror":"https://ror.org/003xyzq10","country_code":"CN","type":"education","lineage":["https://openalex.org/I173899330"]},{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tiehan Cui","raw_affiliation_strings":["School of Artificial Intelligence and Automation, Huazhong University of Science and Technology, Wuhan, China and School of Software, Henan University, Kaifeng, China"],"raw_orcid":"https://orcid.org/0009-0003-7429-8452","affiliations":[{"raw_affiliation_string":"School of Artificial Intelligence and Automation, Huazhong University of Science and Technology, Wuhan, China and School of Software, Henan University, Kaifeng, China","institution_ids":["https://openalex.org/I173899330","https://openalex.org/I47720641"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Peipei Liu","orcid":"https://orcid.org/0000-0003-4621-2298","observed_orcids":["https://orcid.org/0000-0003-4621-2298"]},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210156404","display_name":"Institute of Information Engineering","ror":"https://ror.org/04r53se39","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210156404"]},{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Peipei Liu","raw_affiliation_strings":["Institute of Information Engineering, Chinese Academy of Sciences, Beijing, China and School of Cyberspace Security, University of Chinese Academy of Sciences, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-4621-2298","affiliations":[{"raw_affiliation_string":"Institute of Information Engineering, Chinese Academy of Sciences, Beijing, China and School of Cyberspace Security, University of Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I19820366","https://openalex.org/I4210156404","https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120831784","display_name":"Datao You","orcid":"https://orcid.org/0000-0001-5968-4994","observed_orcids":["https://orcid.org/0000-0001-5968-4994"]},"institutions":[{"id":"https://openalex.org/I173899330","display_name":"Henan University","ror":"https://ror.org/003xyzq10","country_code":"CN","type":"education","lineage":["https://openalex.org/I173899330"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Datao You","raw_affiliation_strings":["School of Software, Henan University, Kaifeng, China"],"raw_orcid":"https://orcid.org/0000-0001-5968-4994","affiliations":[{"raw_affiliation_string":"School of Software, Henan University, Kaifeng, China","institution_ids":["https://openalex.org/I173899330"]}]},{"author_position":"last","author":{"id":null,"display_name":"Hongsong Zhu","orcid":"https://orcid.org/0000-0003-3720-7403","observed_orcids":["https://orcid.org/0000-0003-3720-7403"]},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210156404","display_name":"Institute of Information Engineering","ror":"https://ror.org/04r53se39","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210156404"]},{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hongsong Zhu","raw_affiliation_strings":["Institute of Information Engineering, Chinese Academy of Sciences, Beijing, China and School of Cyberspace Security, University of Chinese Academy of Sciences, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-3720-7403","affiliations":[{"raw_affiliation_string":"Institute of Information Engineering, Chinese Academy of Sciences, Beijing, China and School of Cyberspace Security, University of Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I19820366","https://openalex.org/I4210156404","https://openalex.org/I4210165038"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":20.8834,"has_fulltext":true,"cited_by_count":14,"citation_normalized_percentile":{"value":0.99383537,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"max":99},"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":1.0,"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":1.0,"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/T11424","display_name":"Security and Verification in Computing","score":9.999999747378752e-05,"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/T10028","display_name":"Topic Modeling","score":0.0,"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/jailbreak-attacks","display_name":"jailbreak attacks","score":0.9900000095367432},{"id":"https://openalex.org/keywords/jailbreak-defense","display_name":"jailbreak defense","score":0.9110000133514404},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.9779999852180481},{"id":"https://openalex.org/keywords/multimodal-large-language-models","display_name":"multimodal large language models","score":0.9660000205039978},{"id":"https://openalex.org/keywords/ai-agents","display_name":"AI agents","score":0.7440000176429749},{"id":"https://openalex.org/keywords/llm-security","display_name":"LLM security","score":0.8640000224113464},{"id":"https://openalex.org/keywords/attack-taxonomy","display_name":"attack taxonomy","score":0.6690000295639038},{"id":"https://openalex.org/keywords/evaluation-indicators","display_name":"evaluation metrics","score":0.48899999260902405}],"concepts":[{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5153999924659729},{"id":"https://openalex.org/C2777617010","wikidata":"https://www.wikidata.org/wiki/Q18957","display_name":"Mainstream","level":2,"score":0.47609999775886536},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.43140000104904175},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4194999933242798},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.39410001039505005},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.3752000033855438},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.3634999990463257},{"id":"https://openalex.org/C58642233","wikidata":"https://www.wikidata.org/wiki/Q8269924","display_name":"Taxonomy (biology)","level":2,"score":0.29910001158714294},{"id":"https://openalex.org/C191267431","wikidata":"https://www.wikidata.org/wiki/Q911932","display_name":"Honeypot","level":2,"score":0.29330000281333923},{"id":"https://openalex.org/C539667460","wikidata":"https://www.wikidata.org/wiki/Q2414942","display_name":"Management science","level":1,"score":0.2874999940395355},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.28619998693466187},{"id":"https://openalex.org/C123403432","wikidata":"https://www.wikidata.org/wiki/Q654068","display_name":"Visibility","level":2,"score":0.27900001406669617},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.2784999907016754},{"id":"https://openalex.org/C107826830","wikidata":"https://www.wikidata.org/wiki/Q929380","display_name":"Environmental resource management","level":1,"score":0.26260000467300415},{"id":"https://openalex.org/C195094911","wikidata":"https://www.wikidata.org/wiki/Q14167904","display_name":"Process management","level":1,"score":0.25679999589920044},{"id":"https://openalex.org/C91375879","wikidata":"https://www.wikidata.org/wiki/Q15473274","display_name":"Environmental planning","level":1,"score":0.2538999915122986}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1145/3856301","doi":"https://doi.org/10.1145/3856301","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3856301","pdf_url":null,"source":{"id":"https://openalex.org/S41523882","display_name":"ACM Transactions on Knowledge Discovery from Data","issn_l":"1556-4681","issn":["1556-4681","1556-472X"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["ccf-b","cwts-core","dongbi-a","jufo-3","norway-1","poland-100","russia-white-list-1"],"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Knowledge Discovery from Data","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:2506.15170","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2506.15170","pdf_url":"https://arxiv.org/pdf/2506.15170","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2506.15170","doi":"https://doi.org/10.48550/arxiv.2506.15170","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2506.15170","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":"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2506.15170","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2506.15170","pdf_url":"https://arxiv.org/pdf/2506.15170","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4417423213.pdf","grobid_xml":"https://content.openalex.org/works/W4417423213.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7223563467","https://openalex.org/W7127617783","https://openalex.org/W7131127635","https://openalex.org/W4391709828","https://openalex.org/W4400434624","https://openalex.org/W7168483215","https://openalex.org/W4412944837","https://openalex.org/W7130362751","https://openalex.org/W7160670351","https://openalex.org/W7166851282"],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2],"(LLMs)":[3],"are":[4],"rapidly":[5],"evolving":[6],"from":[7,50,71],"single-modal":[8],"systems":[9],"to":[10,52,126,184],"multimodal":[11],"LLMs":[12,51],"and":[13,36,54,75,79,89,104,146,164,178,191],"intelligent":[14],"agents,":[15],"expanding":[16,41],"their":[17,112],"capabilities":[18],"while":[19],"introducing":[20],"new":[21],"security":[22,59,128],"risks.":[23],"This":[24],"paper":[25],"presents":[26],"a":[27,108,132],"systematic":[28],"survey":[29],"of":[30,33,83,111,131,140,143,149,161,188,195],"the":[31,40,47,57,72,93,186,193],"complexity":[32],"jailbreak":[34,69,137,189],"attacks":[35],"defense":[37,94],"mechanisms":[38,190],"within":[39],"LLM":[42],"ecosystem.":[43],"We":[44],"first":[45],"trace":[46],"developmental":[48],"trajectory":[49],"MLLMs":[53],"Agents,":[55],"highlighting":[56],"core":[58],"challenges":[60],"emerging":[61],"at":[62],"each":[63],"stage.":[64],"Next,":[65],"we":[66,96,115,156],"categorize":[67],"mainstream":[68],"techniques":[70],"attack":[73,85],"impact":[74],"attacker":[76],"permissions":[77],"perspectives,":[78],"provide":[80,157],"comprehensive":[81],"analysis":[82,142],"representative":[84],"methods,":[86,138],"related":[87],"datasets,":[88],"evaluation":[90,175],"metrics.":[91],"On":[92],"side,":[95],"organize":[97],"existing":[98,120],"strategies":[99],"based":[100],"on":[101],"response":[102],"timing":[103],"technical":[105],"approach,":[106],"offering":[107],"structured":[109],"understanding":[110,187],"implementation.":[113],"Furthermore,":[114],"identify":[116],"key":[117],"limitations":[118],"in":[119,169],"surveys,":[121],"such":[122,171],"as":[123,172],"insufficient":[124],"attention":[125],"agent-specific":[127],"issues,":[129],"absence":[130],"clear":[133],"taxonomy":[134],"for":[135,198],"hybrid":[136],"lack":[139],"detailed":[141],"experimental":[144],"setups,":[145],"outdated":[147],"coverage":[148],"recent":[150,162],"advancements.":[151],"To":[152],"address":[153],"these":[154],"limitations,":[155],"an":[158],"updated":[159],"synthesis":[160],"work":[163],"introduce":[165],"future":[166],"research":[167],"directions":[168],"areas":[170],"dataset":[173],"construction,":[174],"framework":[176],"optimization,":[177],"strategy":[179],"generalization.":[180],"Our":[181],"study":[182],"aims":[183],"deepen":[185],"promote":[192],"development":[194],"adaptive":[196],"defenses":[197],"LLMs.":[199]},"counts_by_year":[{"year":2026,"cited_by_count":9},{"year":2025,"cited_by_count":5}],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2025-10-10T00:00:00"}
