{"id":"https://openalex.org/W7131840307","doi":"https://doi.org/10.1109/tcss.2026.3664981","title":"Take off Your Disguise: Detecting Disguised Prompt-Based Jailbreak Attacks Against LLMs","display_name":"Take off Your Disguise: Detecting Disguised Prompt-Based Jailbreak Attacks Against LLMs","publication_year":2026,"publication_date":"2026-02-27","ids":{"openalex":"https://openalex.org/W7131840307","doi":"https://doi.org/10.1109/tcss.2026.3664981"},"language":null,"primary_location":{"id":"doi:10.1109/tcss.2026.3664981","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcss.2026.3664981","pdf_url":null,"source":{"id":"https://openalex.org/S2490693980","display_name":"IEEE Transactions on Computational Social Systems","issn_l":"2329-924X","issn":["2329-924X","2373-7476"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Computational Social Systems","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5127196526","display_name":"Hui Liu","orcid":null},"institutions":[{"id":"https://openalex.org/I40963666","display_name":"Central China Normal University","ror":"https://ror.org/03x1jna21","country_code":"CN","type":"education","lineage":["https://openalex.org/I40963666"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hui Liu","raw_affiliation_strings":["School of Computer Science and Hubei Provincial Key Laboratory of Artificial Intelligence and Smart Learning, Central China Normal University, Wuhan, China"],"raw_orcid":"https://orcid.org/0000-0003-1345-5736","affiliations":[{"raw_affiliation_string":"School of Computer Science and Hubei Provincial Key Laboratory of Artificial Intelligence and Smart Learning, Central China Normal University, Wuhan, China","institution_ids":["https://openalex.org/I40963666"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127418779","display_name":"Fujv Wen","orcid":null},"institutions":[{"id":"https://openalex.org/I40963666","display_name":"Central China Normal University","ror":"https://ror.org/03x1jna21","country_code":"CN","type":"education","lineage":["https://openalex.org/I40963666"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fujv Wen","raw_affiliation_strings":["School of Computer Science and Hubei Provincial Key Laboratory of Artificial Intelligence and Smart Learning, Central China Normal University, Wuhan, China"],"raw_orcid":"https://orcid.org/0009-0009-3898-8003","affiliations":[{"raw_affiliation_string":"School of Computer Science and Hubei Provincial Key Laboratory of Artificial Intelligence and Smart Learning, Central China Normal University, Wuhan, China","institution_ids":["https://openalex.org/I40963666"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5035479831","display_name":"Hongqin Du","orcid":null},"institutions":[{"id":"https://openalex.org/I40963666","display_name":"Central China Normal University","ror":"https://ror.org/03x1jna21","country_code":"CN","type":"education","lineage":["https://openalex.org/I40963666"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hongqin Du","raw_affiliation_strings":["School of Computer Science and Hubei Provincial Key Laboratory of Artificial Intelligence and Smart Learning, Central China Normal University, Wuhan, China"],"raw_orcid":"https://orcid.org/0009-0004-5020-0083","affiliations":[{"raw_affiliation_string":"School of Computer Science and Hubei Provincial Key Laboratory of Artificial Intelligence and Smart Learning, Central China Normal University, Wuhan, China","institution_ids":["https://openalex.org/I40963666"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040336043","display_name":"Jiabao Guo","orcid":"https://orcid.org/0000-0002-2855-4246"},"institutions":[{"id":"https://openalex.org/I16365422","display_name":"Hefei University of Technology","ror":"https://ror.org/02czkny70","country_code":"CN","type":"education","lineage":["https://openalex.org/I16365422"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiabao Guo","raw_affiliation_strings":["School of Computer Science, Hefei University of Technology, Hefei, China"],"raw_orcid":"https://orcid.org/0000-0002-2855-4246","affiliations":[{"raw_affiliation_string":"School of Computer Science, Hefei University of Technology, Hefei, China","institution_ids":["https://openalex.org/I16365422"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5127228586","display_name":"Bo Zhao","orcid":null},"institutions":[{"id":"https://openalex.org/I37461747","display_name":"Wuhan University","ror":"https://ror.org/033vjfk17","country_code":"CN","type":"education","lineage":["https://openalex.org/I37461747"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bo Zhao","raw_affiliation_strings":["School of Cyber Science and Engineering, Wuhan University, Wuhan, China"],"raw_orcid":"https://orcid.org/0000-0003-4307-9380","affiliations":[{"raw_affiliation_string":"School of Cyber Science and Engineering, Wuhan University, Wuhan, China","institution_ids":["https://openalex.org/I37461747"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.16242023,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"13","issue":"3","first_page":"3190","last_page":"3200"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12034","display_name":"Digital and Cyber Forensics","score":0.14010000228881836,"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/T12034","display_name":"Digital and Cyber Forensics","score":0.14010000228881836,"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/T10400","display_name":"Network Security and Intrusion Detection","score":0.13459999859333038,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T11241","display_name":"Advanced Malware Detection Techniques","score":0.11900000274181366,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/adversarial-system","display_name":"Adversarial system","score":0.7297000288963318},{"id":"https://openalex.org/keywords/limiting","display_name":"Limiting","score":0.7235000133514404},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.642799973487854},{"id":"https://openalex.org/keywords/damages","display_name":"Damages","score":0.27810001373291016},{"id":"https://openalex.org/keywords/electronic-mail","display_name":"Electronic mail","score":0.27309998869895935}],"concepts":[{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.7526000142097473},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.7297000288963318},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.7235000133514404},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6976000070571899},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.642799973487854},{"id":"https://openalex.org/C108827166","wikidata":"https://www.wikidata.org/wiki/Q175975","display_name":"Internet privacy","level":1,"score":0.46239998936653137},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33799999952316284},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.3179999887943268},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.2793000042438507},{"id":"https://openalex.org/C2777381055","wikidata":"https://www.wikidata.org/wiki/Q308922","display_name":"Damages","level":2,"score":0.27810001373291016},{"id":"https://openalex.org/C3020028006","wikidata":"https://www.wikidata.org/wiki/Q9158","display_name":"Electronic mail","level":2,"score":0.27309998869895935},{"id":"https://openalex.org/C2778403875","wikidata":"https://www.wikidata.org/wiki/Q20312394","display_name":"Adversarial machine learning","level":3,"score":0.2671000063419342},{"id":"https://openalex.org/C541664917","wikidata":"https://www.wikidata.org/wiki/Q14001","display_name":"Malware","level":2,"score":0.26089999079704285},{"id":"https://openalex.org/C140547941","wikidata":"https://www.wikidata.org/wiki/Q7797194","display_name":"Threat model","level":2,"score":0.2526000142097473}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcss.2026.3664981","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcss.2026.3664981","pdf_url":null,"source":{"id":"https://openalex.org/S2490693980","display_name":"IEEE Transactions on Computational Social Systems","issn_l":"2329-924X","issn":["2329-924X","2373-7476"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Computational Social Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Gender equality","id":"https://metadata.un.org/sdg/5","score":0.622532308101654}],"awards":[{"id":"https://openalex.org/G2138122720","display_name":null,"funder_award_id":"62572356","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G3031900560","display_name":null,"funder_award_id":"2024M751050","funder_id":"https://openalex.org/F4320321543","funder_display_name":"China Postdoctoral Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320321543","display_name":"China Postdoctoral Science Foundation","ror":"https://ror.org/0426zh255"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2],"(LLMs)":[3],"are":[4],"typically":[5],"equipped":[6],"with":[7],"alignment":[8],"mechanisms":[9],"designed":[10,73,90],"to":[11,23,52,74,96,115,169,184],"prevent":[12],"the":[13,24,46,50,94,108,112,119,135,141,162,180],"generation":[14],"of":[15,26,82,111],"harmful":[16],"content.":[17],"However,":[18],"recent":[19],"advances":[20],"have":[21],"led":[22],"emergence":[25],"highly":[27],"evasive":[28],"disguised":[29,98,128],"prompt":[30,61,130],"jailbreak":[31,77,104,129],"attacks":[32],"(DPJAs),":[33],"where":[34],"attackers":[35],"conceal":[36],"real":[37],"malicious":[38,137],"intent":[39,121,138],"within":[40],"prompts":[41,100],"that":[42,154],"appear":[43],"benign":[44],"on":[45],"surface,":[47],"thereby":[48],"inducing":[49],"model":[51],"produce":[53],"unsafe":[54],"outputs.":[55],"In":[56],"this":[57],"work,":[58],"we":[59],"propose":[60],"reconstruction-based":[62],"detection":[63,71],"(<italic":[64],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[65],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">RePrompt</i>),":[66],"a":[67,123,127,173],"training-free":[68],"and":[69,101,117,149,178],"plug-in":[70],"framework":[72],"identify":[75],"such":[76],"attacks.":[78,105],"The":[79],"key":[80],"insight":[81],"RePrompt":[83,106,133,155],"is":[84,131],"that,":[85],"when":[86],"guided":[87],"by":[88],"carefully":[89],"templates,":[91],"LLMs":[92],"possess":[93],"capability":[95],"uncover":[97,118],"adversarial":[99],"effectively":[102],"detect":[103],"leverages":[107],"reasoning":[109],"capabilities":[110],"LLM":[113],"itself":[114],"analyze":[116],"true":[120],"behind":[122],"user\u2019s":[124],"query.":[125],"When":[126],"encountered,":[132],"reconstructs":[134],"underlying":[136],"hidden":[139],"beneath":[140],"surface-level":[142],"disguise.":[143],"Experiments":[144],"across":[145],"multiple":[146],"LLMs,":[147],"datasets,":[148],"three":[150],"representative":[151],"DPJAs":[152],"demonstrate":[153],"consistently":[156],"outperforms":[157],"state-of-the-art":[158],"defense":[159],"methods,":[160],"reducing":[161],"average":[163,181],"attack":[164],"success":[165],"rate":[166],"from":[167],"47.0%":[168],"below":[170],"1.6%,":[171],"achieving":[172],"near-zero":[174],"false":[175],"positive":[176],"rate,":[177],"limiting":[179],"query":[182],"cost":[183],"only":[185],"1.13.":[186]},"counts_by_year":[],"updated_date":"2026-06-02T06:17:35.589633","created_date":"2026-02-28T00:00:00"}