{"id":"https://openalex.org/W4416035875","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.200","title":"Multilingual Collaborative Defense for Large Language Models","display_name":"Multilingual Collaborative Defense for Large Language Models","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416035875","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.200"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.200","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.200","pdf_url":"https://aclanthology.org/2025.findings-emnlp.200.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: EMNLP 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-emnlp.200.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5114378292","display_name":"Hongliang Li","orcid":"https://orcid.org/0000-0002-7481-095X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hongliang Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101698034","display_name":"Jinan Xu","orcid":"https://orcid.org/0000-0003-0170-626X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jinan Xu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Gengping Cui","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gengping Cui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102678874","display_name":"Changhao Guan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Changhao Guan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048033573","display_name":"Fengran Mo","orcid":"https://orcid.org/0000-0002-0838-6994"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fengran Mo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5057728157","display_name":"Kaiyu Huang","orcid":"https://orcid.org/0000-0001-5082-2107"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kaiyu Huang","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":null,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"3735","last_page":"3755"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.2768999934196472,"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":0.2768999934196472,"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/T12262","display_name":"Hate Speech and Cyberbullying Detection","score":0.2485000044107437,"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":0.1673000007867813,"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/natural-language","display_name":"Natural language","score":0.31439998745918274},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.2985999882221222},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.29670000076293945},{"id":"https://openalex.org/keywords/field","display_name":"Field (mathematics)","score":0.2849000096321106},{"id":"https://openalex.org/keywords/multilingualism","display_name":"Multilingualism","score":0.2793000042438507},{"id":"https://openalex.org/keywords/modeling-language","display_name":"Modeling language","score":0.2662999927997589}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6269999742507935},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.4684999883174896},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3935000002384186},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3395000100135803},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.31439998745918274},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2985999882221222},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.29670000076293945},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.2849000096321106},{"id":"https://openalex.org/C136764020","wikidata":"https://www.wikidata.org/wiki/Q466","display_name":"World Wide Web","level":1,"score":0.28040000796318054},{"id":"https://openalex.org/C2780035574","wikidata":"https://www.wikidata.org/wiki/Q30081","display_name":"Multilingualism","level":2,"score":0.2793000042438507},{"id":"https://openalex.org/C179603123","wikidata":"https://www.wikidata.org/wiki/Q1941921","display_name":"Modeling language","level":3,"score":0.2662999927997589},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.258899986743927},{"id":"https://openalex.org/C83479923","wikidata":"https://www.wikidata.org/wiki/Q2063748","display_name":"Universal Networking Language","level":4,"score":0.25540000200271606},{"id":"https://openalex.org/C129792486","wikidata":"https://www.wikidata.org/wiki/Q1050419","display_name":"Language identification","level":3,"score":0.25459998846054077},{"id":"https://openalex.org/C144024400","wikidata":"https://www.wikidata.org/wiki/Q21201","display_name":"Sociology","level":0,"score":0.25440001487731934},{"id":"https://openalex.org/C527412718","wikidata":"https://www.wikidata.org/wiki/Q855395","display_name":"Interpretation (philosophy)","level":2,"score":0.2531000077724457}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.200","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.200","pdf_url":"https://aclanthology.org/2025.findings-emnlp.200.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: EMNLP 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.200","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.200","pdf_url":"https://aclanthology.org/2025.findings-emnlp.200.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: EMNLP 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G106398601","display_name":null,"funder_award_id":"61976015","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G2153459869","display_name":null,"funder_award_id":"61876198","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G310944059","display_name":null,"funder_award_id":"62476023","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G3320668674","display_name":"\u57fa\u4e8e\u683c\u4f9d\u5b58\u6811\u5230\u4e32\u6a21\u578b\u7684\u65e5\u6c49\u673a\u5668\u7ffb\u8bd1\u7814\u7a76","funder_award_id":"61370130","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G5447623188","display_name":null,"funder_award_id":"62376019","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G7838506768","display_name":null,"funder_award_id":"61976016","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320336593","display_name":"Henan Provincial Science and Technology Research Project","ror":null}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416035875.pdf","grobid_xml":"https://content.openalex.org/works/W4416035875.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0],"robustness":[1],"and":[2,82,98,118],"security":[3],"of":[4,58,112],"Large":[5],"Language":[6],"Models":[7],"(LLMs)":[8],"face":[9],"increasing":[10],"threats,":[11],"especially":[12],"in":[13,133],"multilingual":[14,56,110,134],"settings.A":[15],"notable":[16],"vulnerability":[17],"is":[18],"\"jailbreaking\"":[19],"via":[20],"translating":[21],"harmful":[22],"queries":[23],"into":[24],"rare":[25],"or":[26],"underrepresented":[27],"languages,":[28,94,122],"which":[29],"often":[30],"bypasses":[31],"existing":[32],"safeguards.In":[33],"this":[34],"work,":[35],"we":[36,108],"propose":[37],"Multilingual":[38],"Collaborative":[39],"Defense":[40],"(MCD),":[41],"a":[42,48,86],"novel":[43],"learning":[44,81],"method":[45],"that":[46,128],"optimizes":[47],"continuous":[49],"soft":[50],"safety":[51,79,100],"prompt":[52,80],"automatically":[53],"to":[54,77,123],"facilitate":[55],"safeguarding":[57],"LLMs.MCD":[59],"organically":[60],"leverages":[61],"collaborative":[62],"signals":[63],"from":[64],"multiple":[65],"languages":[66,76],"by":[67,103],"rotating":[68],"each":[69],"as":[70,116],"the":[71],"training":[72],"\"center,\"":[73],"allowing":[74],"auxiliary":[75],"reinforce":[78],"ensuring":[83],"cross-lingual":[84],"consistency.As":[85],"result,":[87],"MCD":[88,129],"improves":[89],"defense":[90,136],"performance":[91],"across":[92],"all":[93],"reduces":[95],"false":[96],"refusals,":[97],"mitigates":[99],"misalignment":[101],"caused":[102],"corpus":[104],"imbalance.To":[105],"evaluate":[106],"MCD,":[107],"construct":[109],"versions":[111],"jailbreak":[113,135],"benchmarks":[114],"such":[115],"MaliciousInstruct":[117],"AdvBench,":[119],"including":[120],"zero-shot":[121],"assess":[124],"language":[125],"transferability.Experiments":[126],"show":[127],"outperforms":[130],"prior":[131],"approaches":[132],"while":[137],"exhibiting":[138],"strong":[139],"crosslingual":[140],"generalization.":[141]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-11-08T00:00:00"}