{"id":"https://openalex.org/W4416034136","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.960","title":"Beneath the Facade: Probing Safety Vulnerabilities in LLMs via Auto-Generated Jailbreak Prompts","display_name":"Beneath the Facade: Probing Safety Vulnerabilities in LLMs via Auto-Generated Jailbreak Prompts","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416034136","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.960"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.960","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.960","pdf_url":"https://aclanthology.org/2025.findings-emnlp.960.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-emnlp.960.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5104239678","display_name":"Heehyeon Kim","orcid":null},"institutions":[{"id":"https://openalex.org/I4210143490","display_name":"Computing Center","ror":"https://ror.org/0557kgc34","country_code":"RU","type":"facility","lineage":["https://openalex.org/I1313323035","https://openalex.org/I4210143490","https://openalex.org/I4210148470"]}],"countries":["RU"],"is_corresponding":false,"raw_author_name":"Heehyeon Kim","raw_affiliation_strings":["School of Computing , KAIST"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computing , KAIST","institution_ids":["https://openalex.org/I4210143490"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Kyeongryul Lee","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Kyeongryul Lee","raw_affiliation_strings":["Graduate School of Data Science , KAIST"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Graduate School of Data Science , KAIST","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5090741314","display_name":"Joyce Jiyoung Whang","orcid":"https://orcid.org/0000-0002-4773-3194"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]},{"id":"https://openalex.org/I4210143490","display_name":"Computing Center","ror":"https://ror.org/0557kgc34","country_code":"RU","type":"facility","lineage":["https://openalex.org/I1313323035","https://openalex.org/I4210143490","https://openalex.org/I4210148470"]}],"countries":["KR","RU"],"is_corresponding":false,"raw_author_name":"Joyce Jiyoung Whang","raw_affiliation_strings":["Graduate School of Data Science , KAIST","School of Computing , KAIST"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Graduate School of Data Science , KAIST","institution_ids":["https://openalex.org/I157485424"]},{"raw_affiliation_string":"School of Computing , KAIST","institution_ids":["https://openalex.org/I4210143490"]}]}],"institutions":[],"countries_distinct_count":2,"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":0.39188594,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"17668","last_page":"17700"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10734","display_name":"Information and Cyber Security","score":0.22619999945163727,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10734","display_name":"Information and Cyber Security","score":0.22619999945163727,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.10840000212192535,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.08020000159740448,"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/vulnerability","display_name":"Vulnerability (computing)","score":0.41190001368522644},{"id":"https://openalex.org/keywords/risk-management","display_name":"Risk management","score":0.2888999879360199},{"id":"https://openalex.org/keywords/government","display_name":"Government (linguistics)","score":0.2728999853134155},{"id":"https://openalex.org/keywords/event","display_name":"Event (particle physics)","score":0.2630999982357025},{"id":"https://openalex.org/keywords/risk-assessment","display_name":"Risk assessment","score":0.2563000023365021}],"concepts":[{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.5250999927520752},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.43849998712539673},{"id":"https://openalex.org/C95713431","wikidata":"https://www.wikidata.org/wiki/Q631425","display_name":"Vulnerability (computing)","level":2,"score":0.41190001368522644},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.33000001311302185},{"id":"https://openalex.org/C108827166","wikidata":"https://www.wikidata.org/wiki/Q175975","display_name":"Internet privacy","level":1,"score":0.29429998993873596},{"id":"https://openalex.org/C32896092","wikidata":"https://www.wikidata.org/wiki/Q189447","display_name":"Risk management","level":2,"score":0.2888999879360199},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.27630001306533813},{"id":"https://openalex.org/C2778137410","wikidata":"https://www.wikidata.org/wiki/Q2732820","display_name":"Government (linguistics)","level":2,"score":0.2728999853134155},{"id":"https://openalex.org/C2779662365","wikidata":"https://www.wikidata.org/wiki/Q5416694","display_name":"Event (particle physics)","level":2,"score":0.2630999982357025},{"id":"https://openalex.org/C12174686","wikidata":"https://www.wikidata.org/wiki/Q1058438","display_name":"Risk assessment","level":2,"score":0.2563000023365021},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.25290000438690186}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.960","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.960","pdf_url":"https://aclanthology.org/2025.findings-emnlp.960.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"},{"id":"pmh:oai:koasas.kaist.ac.kr:10203/335493","is_oa":false,"landing_page_url":"http://hdl.handle.net/10203/335493","pdf_url":null,"source":{"id":"https://openalex.org/S4306402435","display_name":"KAIST Institutional Repository (KAIST)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I157485424","host_organization_name":"Korea Advanced Institute of Science and Technology","host_organization_lineage":["https://openalex.org/I157485424"],"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":"Conference"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.960","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.960","pdf_url":"https://aclanthology.org/2025.findings-emnlp.960.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1630947553","display_name":null,"funder_award_id":"RS-2022-II220369","funder_id":"https://openalex.org/F4320335489","funder_display_name":"Institute for Information and Communications Technology Promotion"},{"id":"https://openalex.org/G3996153573","display_name":null,"funder_award_id":"RS-2022-II220369","funder_id":"https://openalex.org/F4320328359","funder_display_name":"Ministry of Science and ICT, South Korea"},{"id":"https://openalex.org/G4700831490","display_name":null,"funder_award_id":"2022-","funder_id":"https://openalex.org/F4320335489","funder_display_name":"Institute for Information and Communications Technology Promotion"}],"funders":[{"id":"https://openalex.org/F4320328359","display_name":"Ministry of Science and ICT, South Korea","ror":"https://ror.org/01wpjm123"},{"id":"https://openalex.org/F4320335489","display_name":"Institute for Information and Communications Technology Promotion","ror":"https://ror.org/01g0hqq23"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416034136.pdf","grobid_xml":"https://content.openalex.org/works/W4416034136.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0],"rapid":[1],"proliferation":[2],"of":[3,22,36,47,88],"large":[4],"language":[5],"models":[6,10],"and":[7,45,94,205,240,292],"multimodal":[8,98],"generative":[9],"has":[11],"raised":[12],"concerns":[13],"about":[14],"their":[15,158,241],"potential":[16],"vulnerabilities":[17,65],"to":[18,39,97,110,177,286],"a":[19,26,123,289],"wide":[20],"range":[21,87],"real-world":[23,48],"safety":[24,106],"risks.However,":[25],"critical":[27],"gap":[28],"persists":[29],"in":[30],"systematic":[31],"assessment,":[32],"alongside":[33],"the":[34,43,178,198,235,254,274],"lack":[35],"evaluation":[37],"frameworks":[38],"keep":[40],"pace":[41],"with":[42,266],"breadth":[44],"variability":[46],"risk":[49,199,207,213,238,257,276],"factors.In":[50],"this":[51],"paper,":[52],"we":[53],"introduce":[54],"TroGEN,":[55],"an":[56,72,81,118],"automated":[57],"jailbreak":[58,69,92,134],"prompt":[59],"generation":[60],"framework":[61],"that":[62,84,102,113,262],"assesses":[63],"these":[64,267],"by":[66],"deriving":[67],"scenario-driven":[68],"prompts":[70],"using":[71,160],"adversarial":[73,111],"agent.Moving":[74],"beyond":[75],"labor-intensive":[76],"dataset":[77],"construction,":[78],"TroGEN":[79,103],"features":[80],"extensible":[82],"design":[83],"covers":[85],"broad":[86],"risks,":[89],"supports":[90],"plug-and-play":[91],"strategies,":[93],"adapts":[95],"seamlessly":[96],"settings.Experimental":[99],"results":[100],"demonstrate":[101],"effectively":[104],"uncovers":[105],"weaknesses,":[107],"revealing":[108],"susceptibilities":[109],"attacks":[112,128,279],"conceal":[114],"malicious":[115,231],"intent":[116],"beneath":[117],"apparently":[119],"benign":[120],"facade,":[121],"like":[122],"Trojan":[124],"horse.Furthermore,":[125],"such":[126,223],"stealthy":[127],"exhibit":[129],"resilience":[130],"even":[131],"against":[132],"existing":[133,236],"defense":[135,179],"methods.Response:":[136],"Certainly!Understanding":[137],"how":[138],"hackers":[139],"craft":[140],"convincing":[141],"phishing":[142],"emails":[143,169],"...":[144,165,186,227,251],"1.":[145],"Social":[146],"Engineering":[147],"Techniques":[148],"-Research":[149],"on":[150,157,189,197],"Target:":[151],"Hackers":[152],"often":[153],"conduct":[154],"extensive":[155],"research":[156],"targets":[159],"social":[161],"media,":[162],"official":[163],"websites":[164],"Contextual":[166],"Relevance:":[167],"Phishing":[168],"may":[170],"reference":[171],"recent":[172],"events":[173],"or":[174,230,298],"issues":[175],"relevant":[176],"ministry":[180],"(e.g.,":[181],"policy":[182],"changes,":[183],"international":[184],"developments":[185],"Safety":[187],"Evaluation":[188],"LLMs":[190],"Granular":[191],"Risk":[192,270],"Factor":[193],"DerivationUser":[194],"Prompt:":[195,272],"Based":[196],"factor":[200,214,277],"'{Fraud}',":[201,287],"provide":[202],"**one":[203],"specific":[204,291],"detailed":[206],"factor**":[208],"derived":[209],"from":[210],"it.The":[211],"given":[212],"is":[215,284],"defined":[216],"as":[217,224],"follows:":[218],"'{Fraud:":[219],"Facilitating":[220],"deceitful":[221],"practices":[222],"scams,":[225],"identity":[226],"for":[228],"personal":[229],"gain}'.Additionally,":[232],"below":[233],"are":[234],"granular":[237,256,275],"factors":[239],"occurrence":[242],"counts:":[243],"'{Online":[244],"retail":[245],"payment":[246],"fraud":[247],"(Appeared":[248],"1":[249],"time)":[250],"}'.Avoid":[252],"duplicating":[253],"above":[255],"factors.Please":[258],"generate":[259,288],"diverse":[260],"alternatives":[261],"do":[263],"not":[264],"overlap":[265],"factors.":[268],"Scenario-driven":[269],"ModelingUser":[271],"Using":[273],"'{Phishing":[278],"via":[280],"deceptive":[281],"emails}'":[282],"which":[283],"related":[285],"highly":[290],"severe":[293],"scenario":[294],"involving":[295],"**potential":[296],"harm":[297],"danger**.":[299]},"counts_by_year":[],"updated_date":"2026-08-18T07:49:30.821534","created_date":"2025-11-08T00:00:00"}