{"id":"https://openalex.org/W4416018081","doi":"https://doi.org/10.1145/3746252.3760899","title":"Jailbreaking LLMs Through Alignment Vulnerabilities in Out-of-Distribution Settings","display_name":"Jailbreaking LLMs Through Alignment Vulnerabilities in Out-of-Distribution Settings","publication_year":2025,"publication_date":"2025-11-08","ids":{"openalex":"https://openalex.org/W4416018081","doi":"https://doi.org/10.1145/3746252.3760899"},"language":null,"primary_location":{"id":"doi:10.1145/3746252.3760899","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746252.3760899","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 34th ACM International Conference on Information and Knowledge Management","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3746252.3760899","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5103549861","display_name":"Yue Huang","orcid":"https://orcid.org/0000-0002-8315-7972"},"institutions":[{"id":"https://openalex.org/I107639228","display_name":"University of Notre Dame","ror":"https://ror.org/00mkhxb43","country_code":"US","type":"education","lineage":["https://openalex.org/I107639228"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yue Huang","raw_affiliation_strings":["University of Notre Dame, South Bend, IN, USA"],"raw_orcid":"https://orcid.org/0000-0002-8315-7972","affiliations":[{"raw_affiliation_string":"University of Notre Dame, South Bend, IN, USA","institution_ids":["https://openalex.org/I107639228"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101384273","display_name":"Jingyu Tang","orcid":"https://orcid.org/0009-0008-8259-8531"},"institutions":[{"id":"https://openalex.org/I107639228","display_name":"University of Notre Dame","ror":"https://ror.org/00mkhxb43","country_code":"US","type":"education","lineage":["https://openalex.org/I107639228"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jingyu Tang","raw_affiliation_strings":["University of Notre Dame, South Bend, IN, USA"],"raw_orcid":"https://orcid.org/0009-0008-8259-8531","affiliations":[{"raw_affiliation_string":"University of Notre Dame, South Bend, IN, USA","institution_ids":["https://openalex.org/I107639228"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015370659","display_name":"Dongping Chen","orcid":"https://orcid.org/0009-0009-9848-2557"},"institutions":[{"id":"https://openalex.org/I66946132","display_name":"University of Maryland, College Park","ror":"https://ror.org/047s2c258","country_code":"US","type":"education","lineage":["https://openalex.org/I66946132"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Dongping Chen","raw_affiliation_strings":["University of Maryland, College Park, MD, USA"],"raw_orcid":"https://orcid.org/0009-0009-9848-2557","affiliations":[{"raw_affiliation_string":"University of Maryland, College Park, MD, USA","institution_ids":["https://openalex.org/I66946132"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101389075","display_name":"Bingda Tang","orcid":"https://orcid.org/0009-0003-2522-6858"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bingda Tang","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0003-2522-6858","affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101555948","display_name":"Yao Wan","orcid":"https://orcid.org/0000-0001-6937-4180"},"institutions":[{"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":"Yao Wan","raw_affiliation_strings":["Huazhong University of Science and Technology, Wuhan, China"],"raw_orcid":"https://orcid.org/0000-0001-6937-4180","affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology, Wuhan, China","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015105117","display_name":"Lichao Sun","orcid":"https://orcid.org/0000-0003-1539-7939"},"institutions":[{"id":"https://openalex.org/I186143895","display_name":"Lehigh University","ror":"https://ror.org/012afjb06","country_code":"US","type":"education","lineage":["https://openalex.org/I186143895"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Lichao Sun","raw_affiliation_strings":["Lehigh University, Bethlehem, PA, USA"],"raw_orcid":"https://orcid.org/0000-0003-1539-7939","affiliations":[{"raw_affiliation_string":"Lehigh University, Bethlehem, PA, USA","institution_ids":["https://openalex.org/I186143895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036357902","display_name":"Philip S. Yu","orcid":"https://orcid.org/0000-0002-3491-5968"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]},{"id":"https://openalex.org/I39422238","display_name":"University of Illinois Chicago","ror":"https://ror.org/02mpq6x41","country_code":"US","type":"education","lineage":["https://openalex.org/I39422238"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Philip Yu","raw_affiliation_strings":["University of Illinois Chicago, Chicgao, IL, USA"],"raw_orcid":"https://orcid.org/0000-0002-3491-5968","affiliations":[{"raw_affiliation_string":"University of Illinois Chicago, Chicgao, IL, USA","institution_ids":["https://openalex.org/I157725225","https://openalex.org/I39422238"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5000755750","display_name":"Xiangliang Zhang","orcid":"https://orcid.org/0000-0002-3574-5665"},"institutions":[{"id":"https://openalex.org/I107639228","display_name":"University of Notre Dame","ror":"https://ror.org/00mkhxb43","country_code":"US","type":"education","lineage":["https://openalex.org/I107639228"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xiangliang Zhang","raw_affiliation_strings":["University of Notre Dame, South Bend, IN, USA"],"raw_orcid":"https://orcid.org/0000-0002-3574-5665","affiliations":[{"raw_affiliation_string":"University of Notre Dame, South Bend, IN, USA","institution_ids":["https://openalex.org/I107639228"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":8,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.7588,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.89377772,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"4817","last_page":"4821"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.8215000033378601,"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":0.8215000033378601,"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":0.050200000405311584,"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":0.01769999973475933,"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/limiting","display_name":"Limiting","score":0.5724999904632568},{"id":"https://openalex.org/keywords/simple","display_name":"Simple (philosophy)","score":0.44200000166893005},{"id":"https://openalex.org/keywords/work","display_name":"Work (physics)","score":0.43230000138282776},{"id":"https://openalex.org/keywords/vulnerability","display_name":"Vulnerability (computing)","score":0.37380000948905945},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.367000013589859}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6018999814987183},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.5724999904632568},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.5407999753952026},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.451200008392334},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.44200000166893005},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.43230000138282776},{"id":"https://openalex.org/C95713431","wikidata":"https://www.wikidata.org/wiki/Q631425","display_name":"Vulnerability (computing)","level":2,"score":0.37380000948905945},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.367000013589859},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.31630000472068787},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.31520000100135803},{"id":"https://openalex.org/C108827166","wikidata":"https://www.wikidata.org/wiki/Q175975","display_name":"Internet privacy","level":1,"score":0.296099990606308},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.27950000762939453},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.27230000495910645},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.258899986743927},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2565999925136566}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3746252.3760899","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746252.3760899","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 34th ACM International Conference on Information and Knowledge Management","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3746252.3760899","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746252.3760899","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 34th ACM International Conference on Information and Knowledge Management","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":3,"referenced_works":["https://openalex.org/W4396843999","https://openalex.org/W4402671039","https://openalex.org/W4410609100"],"related_works":[],"abstract_inverted_index":{"Recently,":[0],"Large":[1],"Language":[2],"Models":[3],"(LLMs)":[4],"have":[5],"shown":[6],"remarkable":[7],"capabilities,":[8],"but":[9],"concerns":[10],"about":[11],"their":[12],"trustworthiness-especially":[13],"under":[14],"''jailbreaking''":[15],"attacks-remain":[16],"unresolved.":[17],"Prior":[18],"work":[19],"often":[20],"assumes":[21],"white-box":[22],"access":[23],"or":[24],"relies":[25],"on":[26,47],"fixed":[27],"prompt":[28],"templates,":[29],"limiting":[30],"practicality.":[31],"We":[32],"propose":[33],"ObscurePrompt,":[34],"a":[35],"simple":[36],"yet":[37],"effective":[38,81],"black-box":[39],"jailbreak":[40,57],"method":[41],"inspired":[42],"by":[43],"fragile":[44],"LLM":[45],"alignment":[46],"Out-of-Distribution":[48],"(OOD)":[49],"inputs.":[50],"ObscurePrompt":[51,75],"constructs":[52],"base":[53],"prompts":[54],"using":[55],"existing":[56,77],"techniques,":[58],"then":[59],"employs":[60],"powerful":[61],"LLMs":[62],"to":[63],"iteratively":[64],"generate":[65],"obscure":[66],"variants":[67],"that":[68,74],"evade":[69],"detection.":[70],"Extensive":[71],"experiments":[72],"demonstrate":[73],"outperforms":[76],"methods":[78],"and":[79],"remains":[80],"against":[82],"two":[83],"widely-used":[84],"defenses.":[85]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-11-08T00:00:00"}