{"id":"https://openalex.org/W4412888273","doi":"https://doi.org/10.18653/v1/2025.findings-acl.721","title":"SafeLawBench: Towards Safe Alignment of Large Language Models","display_name":"SafeLawBench: Towards Safe Alignment of Large Language Models","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412888273","doi":"https://doi.org/10.18653/v1/2025.findings-acl.721"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.721","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.721","pdf_url":"https://aclanthology.org/2025.findings-acl.721.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":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.721.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Chuxue Cao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chuxue Cao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023381631","display_name":"Zhu Han","orcid":"https://orcid.org/0000-0002-8602-864X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Han Zhu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5010983432","display_name":"Jiaming Ji","orcid":"https://orcid.org/0000-0002-3769-2077"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiaming Ji","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Qichao Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qichao Sun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5086357316","display_name":"Zining Zhu","orcid":"https://orcid.org/0000-0002-9746-9977"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhenghao Zhu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119181393","display_name":"Wu Yinyu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu Yinyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5099380859","display_name":"Josef Dai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Josef Dai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090073634","display_name":"Yaodong Yang","orcid":"https://orcid.org/0000-0001-8132-5613"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yaodong Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077447568","display_name":"Sirui Han","orcid":"https://orcid.org/0000-0001-9461-8538"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sirui Han","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5045081171","display_name":"Yike Guo","orcid":"https://orcid.org/0000-0002-3075-2161"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yike Guo","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":4,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"14015","last_page":"14048"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.7838000059127808,"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/T10028","display_name":"Topic Modeling","score":0.7838000059127808,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.7490000128746033,"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/computer-science","display_name":"Computer science","score":0.7291643619537354},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.3863193392753601},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.3608272671699524}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7291643619537354},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.3863193392753601},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3608272671699524}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.721","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.721","pdf_url":"https://aclanthology.org/2025.findings-acl.721.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"},{"id":"pmh:oai:repository.hkust.edu.hk:1783.1-149425","is_oa":false,"landing_page_url":"http://repository.hkust.edu.hk/ir/Record/1783.1-149425","pdf_url":null,"source":{"id":"https://openalex.org/S4306401796","display_name":"Rare & Special e-Zone (The Hong Kong University of Science and Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I200769079","host_organization_name":"Hong Kong University of Science and Technology","host_organization_lineage":["https://openalex.org/I200769079"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Conference paper"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.721","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.721","pdf_url":"https://aclanthology.org/2025.findings-acl.721.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":0.5799999833106995,"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities"}],"awards":[{"id":"https://openalex.org/G1142957533","display_name":null,"funder_award_id":"R9911","funder_id":"https://openalex.org/F4320323537","funder_display_name":"Hong Kong University of Science and Technology"}],"funders":[{"id":"https://openalex.org/F4320323537","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412888273.pdf","grobid_xml":"https://content.openalex.org/works/W4412888273.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/W4396696052"],"abstract_inverted_index":{"With":[0],"the":[1,9,31,43,56,102,110,147,157,163],"growing":[2],"prevalence":[3],"of":[4,11,22,34,46,105,150,165],"large":[5],"language":[6],"models":[7,132],"(LLMs),":[8],"safety":[10,28,36,48,60,103,164],"LLMs":[12,91,95,152],"has":[13],"raised":[14],"significant":[15],"concerns.However,":[16],"there":[17],"is":[18],"still":[19],"a":[20,51,70,121],"lack":[21],"definitive":[23],"standards":[24],"for":[25,75],"evaluating":[26],"their":[27],"due":[29],"to":[30,159],"subjective":[32],"nature":[33],"current":[35],"benchmarks.To":[37],"address":[38],"this":[39],"gap,":[40],"we":[41,118],"conducted":[42],"first":[44],"exploration":[45],"LLMs'":[47,111],"evaluation":[49,87],"from":[50],"legal":[52,67],"perspective":[53],"by":[54],"proposing":[55],"SafeLawBench":[57],"benchmark.SafeLawBench":[58],"categorizes":[59],"risks":[61],"into":[62],"three":[63],"levels":[64],"based":[65],"on":[66,144,162],"standards,":[68],"providing":[69],"systematic":[71],"and":[72,81,92,98,115,168],"comprehensive":[73],"framework":[74],"evaluation.It":[76],"comprises":[77],"24,860":[78],"multi-choice":[79,142],"questions":[80],"1,106":[82],"open-domain":[83],"question-answering":[84],"(QA)":[85],"tasks.Our":[86],"included":[88],"2":[89],"closed-source":[90],"18":[93],"open-source":[94],"using":[96],"zero-shot":[97],"fewshot":[99],"prompting,":[100],"highlighting":[101],"features":[104],"each":[106],"model.We":[107],"also":[108],"evaluated":[109],"safety-related":[112],"reasoning":[113],"stability":[114],"refusal":[116],"behavior.Additionally,":[117],"found":[119],"that":[120],"majority":[122],"voting":[123],"mechanism":[124],"can":[125],"enhance":[126],"model":[127],"performance.Notably,":[128],"even":[129],"leading":[130],"SOTA":[131],"like":[133],"Claude-3.5-Sonnetand":[134],"GPT-4o":[135],"have":[136],"not":[137],"exceeded":[138],"80.5%":[139],"accuracy":[140,149],"in":[141],"tasks":[143],"SafeLawBench,":[145],"while":[146],"average":[148],"20":[151],"remains":[153],"at":[154],"68.8%.We":[155],"urge":[156],"community":[158],"prioritize":[160],"research":[161],"LLMs.Our":[166],"dataset":[167],"code":[169],"are":[170],"available.":[171]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}