{"id":"https://openalex.org/W7166808473","doi":"https://doi.org/10.18653/v1/2026.findings-acl.355","title":"Three Minds, One Legend: Jailbreak Large Reasoning Model with Adaptive Stacked Ciphers","display_name":"Three Minds, One Legend: Jailbreak Large Reasoning Model with Adaptive Stacked Ciphers","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166808473","doi":"https://doi.org/10.18653/v1/2026.findings-acl.355"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.355","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.355","pdf_url":"https://aclanthology.org/2026.findings-acl.355.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: ACL 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.355.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139752233","display_name":"Nguyen Viet Anh","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nguyen Viet Anh","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073951885","display_name":"Shiqian Zhao","orcid":"https://orcid.org/0000-0002-4826-1627"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shiqian Zhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139817001","display_name":"Gia Dao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gia Dao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101285238","display_name":"Runyi Hu","orcid":"https://orcid.org/0009-0001-6974-2542"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Runyi Hu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139720498","display_name":"Yi Xie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yi Xie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139715013","display_name":"Xiaobao Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiaobao Wu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5050386762","display_name":"Anh Tuan Luu","orcid":"https://orcid.org/0000-0002-1927-9895"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Anh Tuan Luu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"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":0.93044933,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"7139","last_page":"7159"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12572","display_name":"Intelligence, Security, War Strategy","score":0.12229999899864197,"subfield":{"id":"https://openalex.org/subfields/3320","display_name":"Political Science and International Relations"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T12572","display_name":"Intelligence, Security, War Strategy","score":0.12229999899864197,"subfield":{"id":"https://openalex.org/subfields/3320","display_name":"Political Science and International Relations"},"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/T12002","display_name":"Computability, Logic, AI Algorithms","score":0.07410000264644623,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T10951","display_name":"Cryptographic Implementations and Security","score":0.05900000035762787,"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/key","display_name":"Key (lock)","score":0.2777999937534332},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.26739999651908875},{"id":"https://openalex.org/keywords/scheme","display_name":"Scheme (mathematics)","score":0.2590999901294708},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.24959999322891235}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6128000020980835},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4440999925136566},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38100001215934753},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.34599998593330383},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2777999937534332},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.26739999651908875},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.2590999901294708},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.24959999322891235},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.24729999899864197},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.23510000109672546}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.355","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.355","pdf_url":"https://aclanthology.org/2026.findings-acl.355.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: ACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.355","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.355","pdf_url":"https://aclanthology.org/2026.findings-acl.355.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: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320320671","display_name":"National Research Foundation","ror":"https://ror.org/05s0g1g46"},{"id":"https://openalex.org/F4320320709","display_name":"National Research Foundation Singapore","ror":"https://ror.org/03cpyc314"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166808473.pdf","grobid_xml":"https://content.openalex.org/works/W7166808473.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recently,":[0],"Large":[1,13],"Reasoning":[2],"Models":[3,15],"(LRMs)":[4],"have":[5],"demonstrated":[6],"superior":[7],"logical":[8],"capabilities":[9],"compared":[10],"to":[11,29,42,69,89],"traditional":[12],"Language":[14],"(LLMs),":[16],"gaining":[17],"significant":[18,156],"attention.Despite":[19],"their":[20,71],"impressive":[21],"performance,":[22],"the":[23,91,116,135],"potential":[24,76],"for":[25],"stronger":[26],"reasoning":[27,72,93,125],"abilities":[28],"introduce":[30],"more":[31],"severe":[32],"security":[33],"vulnerabilities":[34],"remains":[35],"largely":[36],"underexplored.Existing":[37],"jailbreak":[38,58],"methods":[39],"often":[40],"struggle":[41],"balance":[43],"effectiveness":[44,136],"with":[45],"robustness":[46],"against":[47],"adaptive":[48,65,77],"safety":[49,98],"mechanisms.In":[50],"this":[51],"work,":[52],"we":[53,106],"propose":[54],"SEAL,":[55],"a":[56,81,155],"novel":[57],"attack":[59,143],"that":[60,85,114],"targets":[61],"LRMs":[62,102],"through":[63],"an":[64,142],"encryption":[66,83],"pipeline":[67],"designed":[68],"override":[70],"processes":[73],"and":[74,112,120,131,167],"evade":[75],"alignment.Specifically,":[78],"SEAL":[79,140],"introduces":[80],"stacked":[82],"approach":[84],"combines":[86],"multiple":[87],"ciphers":[88],"overwhelm":[90],"model's":[92],"capabilities,":[94],"effectively":[95],"bypassing":[96],"builtin":[97],"mechanisms.To":[99],"further":[100],"prevent":[101],"from":[103],"developing":[104],"countermeasures,":[105],"incorporate":[107],"two":[108],"dynamic":[109],"strategies,":[110],"random":[111],"adaptive,":[113],"adjust":[115],"cipher":[117],"length,":[118],"order,":[119],"combination.Extensive":[121],"experiments":[122],"on":[123,148],"real-world":[124],"models,":[126],"including":[127],"DeepSeek-R1,":[128],"Claude":[129],"Sonnet,":[130],"Ope-nAI":[132],"GPT-o4-mini,":[133],"validate":[134],"of":[137,146,158,164],"our":[138],"approach.Notably,":[139],"achieves":[141],"success":[144],"rate":[145],"85.6%":[147],"GPT":[149],"o4-mini,":[150],"outperforming":[151],"state-of-the-art":[152],"baselines":[153],"by":[154],"margin":[157],"17.2%.Warning:":[159],"This":[160],"paper":[161],"contains":[162],"examples":[163],"inappropriate,":[165],"offensive,":[166],"harmful":[168],"content.":[169]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}