{"id":"https://openalex.org/W4412888479","doi":"https://doi.org/10.18653/v1/2025.findings-acl.408","title":"A Mousetrap: Fooling Large Reasoning Models for Jailbreak with Chain of Iterative Chaos","display_name":"A Mousetrap: Fooling Large Reasoning Models for Jailbreak with Chain of Iterative Chaos","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412888479","doi":"https://doi.org/10.18653/v1/2025.findings-acl.408"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.408","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.408","pdf_url":"https://aclanthology.org/2025.findings-acl.408.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 2025","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.408.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5023082475","display_name":"Yao Yang","orcid":"https://orcid.org/0000-0002-7039-4848"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang Yao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101386245","display_name":"Xuan Tong","orcid":"https://orcid.org/0009-0001-4379-4504"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xuan Tong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5080373487","display_name":"Ruofan Wang","orcid":"https://orcid.org/0009-0005-0949-9607"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ruofan Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059464470","display_name":"Yixu Wang","orcid":"https://orcid.org/0009-0008-5104-312X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yixu Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008680227","display_name":"Leya Li","orcid":"https://orcid.org/0009-0001-6533-5034"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lujundong Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069776729","display_name":"Liang Liu","orcid":"https://orcid.org/0000-0003-3794-0891"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liang Liu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100525584","display_name":"Yan Teng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yan Teng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100613148","display_name":"Yingchun Wang","orcid":"https://orcid.org/0000-0003-4968-8398"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yingchun Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":8,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.7588,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.87768398,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"7837","last_page":"7855"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11574","display_name":"Artificial Intelligence in Games","score":0.8780999779701233,"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/T11574","display_name":"Artificial Intelligence in Games","score":0.8780999779701233,"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/T12002","display_name":"Computability, Logic, AI Algorithms","score":0.8628000020980835,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/chaos","display_name":"CHAOS (operating system)","score":0.764649510383606},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6481804251670837},{"id":"https://openalex.org/keywords/chain","display_name":"Chain (unit)","score":0.42529869079589844},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.37077009677886963},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.08371526002883911},{"id":"https://openalex.org/keywords/computer-security","display_name":"Computer security","score":0.06681603193283081}],"concepts":[{"id":"https://openalex.org/C2779374083","wikidata":"https://www.wikidata.org/wiki/Q5011038","display_name":"CHAOS (operating system)","level":2,"score":0.764649510383606},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6481804251670837},{"id":"https://openalex.org/C199185054","wikidata":"https://www.wikidata.org/wiki/Q552299","display_name":"Chain (unit)","level":2,"score":0.42529869079589844},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.37077009677886963},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.08371526002883911},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.06681603193283081},{"id":"https://openalex.org/C1276947","wikidata":"https://www.wikidata.org/wiki/Q333","display_name":"Astronomy","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.408","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.408","pdf_url":"https://aclanthology.org/2025.findings-acl.408.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 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.408","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.408","pdf_url":"https://aclanthology.org/2025.findings-acl.408.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 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.4300000071525574,"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412888479.pdf","grobid_xml":"https://content.openalex.org/works/W4412888479.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W2320196587"],"abstract_inverted_index":{"Large":[0,9],"Reasoning":[1],"Models":[2,11],"(LRMs)":[3],"have":[4],"significantly":[5],"advanced":[6,84],"beyond":[7],"traditional":[8],"Language":[10],"(LLMs)":[12],"with":[13,99,145],"their":[14,30,78],"exceptional":[15],"logical":[16],"reasoning":[17,48,63,85,114,163],"capabilities,":[18],"yet":[19],"these":[20],"improvements":[21],"introduce":[22,88],"heightened":[23],"safety":[24],"risks.When":[25],"subjected":[26],"to":[27,32,41,95,150],"jailbreak":[28,73],"attacks,":[29,55],"ability":[31],"generate":[33],"more":[34,126,152],"targeted":[35],"and":[36,120,122,164,176,184,197,214,222],"organized":[37],"content":[38],"can":[39,206],"lead":[40],"greater":[42],"harm.Although":[43],"some":[44],"studies":[45],"claim":[46],"that":[47],"enables":[49],"safer":[50],"LRMs":[51,155],"against":[52],"existing":[53],"LLM":[54],"they":[56],"overlook":[57],"the":[58,62,71,83,108,113,118,133,151,158,171],"inherent":[59],"flaws":[60],"within":[61],"process":[64],"itself.To":[65],"address":[66],"this":[67],"gap,":[68],"we":[69,87,131],"propose":[70],"first":[72],"attack":[74,97],"targeting":[75],"LRMs,":[76],"exploiting":[77],"unique":[79],"vulnerabilities":[80],"stemming":[81],"from":[82],"capabilities.Specifically,":[86],"a":[89,92,125],"CHAOS":[90],"MACHINE,":[91],"novel":[93],"component":[94],"transform":[96],"prompts":[98],"diverse":[100],"one-to-one":[101],"mappings.The":[102],"chaos":[103],"mappings":[104],"iteratively":[105],"generated":[106],"by":[107],"machine":[109],"are":[110,178],"embedded":[111],"into":[112,140,166],"chain,":[115],"which":[116,136],"strengthens":[117],"variability":[119],"complexity":[121],"also":[123],"promotes":[124],"robust":[127],"attack.Based":[128],"on":[129,187],"this,":[130],"construct":[132],"MOUSETRAP":[134],"framework,":[135],"makes":[137],"attacks":[138],"projected":[139],"nonlinearlike":[141],"low":[142],"sample":[143],"spaces":[144],"mismatched":[146],"generalization":[147],"enhanced.Also,":[148],"due":[149],"competing":[153],"objectives,":[154],"gradually":[156],"maintain":[157],"inertia":[159],"of":[160,170,211],"unpredictable":[161],"iterative":[162],"fall":[165],"our":[167,188],"trap.Success":[168],"rates":[169,210],"Mousetrap":[172,205],"attacking":[173,199],"o1-mini,":[174],"Claude-Sonnet":[175],"Gemini-Thinking":[177],"as":[179,181,194],"high":[180],"96%,":[182],"86%":[183],"98%":[185],"respectively":[186],"toxic":[189],"dataset":[190],"TROTTER.On":[191],"benchmarks":[192],"such":[193],"AdvBench,":[195],"StrongREJECT,":[196],"HarmBench,":[198],"Claude-Sonnet,":[200],"well-known":[201],"for":[202],"its":[203],"safety,":[204],"astonishingly":[207],"achieve":[208],"success":[209],"87.5%,":[212],"86.58%":[213],"93.13%":[215],"respectively.Attention:":[216],"This":[217],"paper":[218],"contains":[219],"inappropriate,":[220],"offensive":[221],"harmful":[223],"content.":[224]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}
