{"id":"https://openalex.org/W4412889939","doi":"https://doi.org/10.18653/v1/2025.acl-long.1282","title":"MTSA: Multi-turn Safety Alignment for LLMs through Multi-round Red-teaming","display_name":"MTSA: Multi-turn Safety Alignment for LLMs through Multi-round Red-teaming","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412889939","doi":"https://doi.org/10.18653/v1/2025.acl-long.1282"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.1282","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1282","pdf_url":"https://aclanthology.org/2025.acl-long.1282.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":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.acl-long.1282.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5102723985","display_name":"Wei Guo","orcid":"https://orcid.org/0000-0003-4257-1453"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Weiyang Guo","raw_affiliation_strings":["Harbin Institute of Technology , Shenzhen , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology , Shenzhen , China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100646613","display_name":"Jing Li","orcid":"https://orcid.org/0000-0003-3027-4355"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jing Li","raw_affiliation_strings":["Harbin Institute of Technology , Shenzhen , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology , Shenzhen , China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101936536","display_name":"Wenya Wang","orcid":"https://orcid.org/0000-0003-2624-5498"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Wenya Wang","raw_affiliation_strings":["Nanyang Technological University , Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanyang Technological University , Singapore","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100702455","display_name":"Yu Li","orcid":"https://orcid.org/0000-0003-2694-5099"},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yu Li","raw_affiliation_strings":["Zhejiang University , Zhejiang , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University , Zhejiang , China","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082144617","display_name":"Daojing He","orcid":"https://orcid.org/0000-0002-3820-8128"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Daojing He","raw_affiliation_strings":["Harbin Institute of Technology , Shenzhen , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology , Shenzhen , China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054413430","display_name":"Jun Yu","orcid":"https://orcid.org/0000-0003-0575-9334"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jun Yu","raw_affiliation_strings":["Harbin Institute of Technology , Shenzhen , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology , Shenzhen , China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5102799323","display_name":"Min Zhang","orcid":"https://orcid.org/0009-0008-6037-529X"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Min Zhang","raw_affiliation_strings":["Harbin Institute of Technology , Shenzhen , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology , Shenzhen , China","institution_ids":["https://openalex.org/I204983213"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":12.7249,"has_fulltext":true,"cited_by_count":4,"citation_normalized_percentile":{"value":0.98842225,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"26424","last_page":"26442"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T13923","display_name":"Nuclear and radioactivity studies","score":0.7409999966621399,"subfield":{"id":"https://openalex.org/subfields/2213","display_name":"Safety, Risk, Reliability and Quality"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T13923","display_name":"Nuclear and radioactivity studies","score":0.7409999966621399,"subfield":{"id":"https://openalex.org/subfields/2213","display_name":"Safety, Risk, Reliability and Quality"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11242","display_name":"Nuclear Materials and Properties","score":0.7142999768257141,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11357","display_name":"Risk and Safety Analysis","score":0.6973000168800354,"subfield":{"id":"https://openalex.org/subfields/1804","display_name":"Statistics, Probability and Uncertainty"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/turn","display_name":"Turn (biochemistry)","score":0.647844135761261},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.48647305369377136},{"id":"https://openalex.org/keywords/chemistry","display_name":"Chemistry","score":0.11796560883522034}],"concepts":[{"id":"https://openalex.org/C85641259","wikidata":"https://www.wikidata.org/wiki/Q290042","display_name":"Turn (biochemistry)","level":2,"score":0.647844135761261},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.48647305369377136},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.11796560883522034},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.1282","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1282","pdf_url":"https://aclanthology.org/2025.acl-long.1282.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":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.acl-long.1282","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1282","pdf_url":"https://aclanthology.org/2025.acl-long.1282.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":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G6570941292","display_name":null,"funder_award_id":"62125201","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412889939.pdf","grobid_xml":"https://content.openalex.org/works/W4412889939.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/W4412839308","https://openalex.org/W4231070408","https://openalex.org/W943151747","https://openalex.org/W2048981943","https://openalex.org/W2753901553","https://openalex.org/W2804368879","https://openalex.org/W4245567755"],"abstract_inverted_index":{"The":[0],"proliferation":[1],"of":[2,51,58,114],"jailbreak":[3,74],"attacks":[4,75],"against":[5],"large":[6],"language":[7],"models":[8],"(LLMs)":[9],"highlights":[10],"the":[11,41,49,62,67,80,85,89,112,120,128],"need":[12],"for":[13],"robust":[14],"security":[15],"measures.However,":[16],"in":[17,25,54,97],"multi-round":[18,55],"dialogues,":[19],"malicious":[20],"intentions":[21],"may":[22],"be":[23,30],"hidden":[24],"interactions,":[26],"leading":[27],"LLMs":[28,53],"to":[29,33,47,76,110],"more":[31],"prone":[32],"produce":[34],"harmful":[35],"responses.In":[36],"this":[37],"paper,":[38],"we":[39,99],"propose":[40],"Multi-Turn":[42],"Safety":[43],"Alignment":[44],"(MTSA)":[45],"framework,":[46],"address":[48],"challenge":[50],"securing":[52],"interactions.It":[56],"consists":[57],"two":[59],"stages:":[60],"In":[61],"thought-guided":[63,72],"attack":[64,125],"learning":[65,104],"stage,":[66,84],"redteam":[68],"model":[69,87,91,122,130],"learns":[70],"about":[71],"multiround":[73],"generate":[77],"adversarial":[78,81],"prompts.In":[79],"iterative":[82],"optimization":[83],"red-team":[86,121],"and":[88],"target":[90,129],"continuously":[92],"improve":[93],"their":[94],"respective":[95],"capabilities":[96],"interaction.Furthermore,":[98],"introduce":[100],"a":[101],"multi-turn":[102],"reinforcement":[103],"algorithm":[105],"based":[106],"on":[107,135],"future":[108],"rewards":[109],"enhance":[111],"robustness":[113],"safety":[115,136],"alignment.Experimental":[116],"results":[117],"show":[118],"that":[119],"exhibits":[123],"state-of-the-art":[124],"capabilities,":[126],"while":[127],"significantly":[131],"improves":[132],"its":[133],"performance":[134],"benchmarks.":[137]},"counts_by_year":[{"year":2026,"cited_by_count":4}],"updated_date":"2026-08-21T09:56:20.448147","created_date":"2025-10-10T00:00:00"}