{"id":"https://openalex.org/W4416035991","doi":"https://doi.org/10.18653/v1/2025.emnlp-main.1091","title":"Interpretation Meets Safety: A Survey on Interpretation Methods and Tools for Improving LLM Safety","display_name":"Interpretation Meets Safety: A Survey on Interpretation Methods and Tools for Improving LLM Safety","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416035991","doi":"https://doi.org/10.18653/v1/2025.emnlp-main.1091"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.emnlp-main.1091","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.1091","pdf_url":"https://aclanthology.org/2025.emnlp-main.1091.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 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.emnlp-main.1091.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101442992","display_name":"Seongmin Lee","orcid":"https://orcid.org/0000-0002-1950-5004"},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Seongmin Lee","raw_affiliation_strings":["Georgia Tech"],"affiliations":[{"raw_affiliation_string":"Georgia Tech","institution_ids":["https://openalex.org/I130701444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119876028","display_name":"Aeree Cho","orcid":null},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Aeree Cho","raw_affiliation_strings":["Georgia Tech"],"affiliations":[{"raw_affiliation_string":"Georgia Tech","institution_ids":["https://openalex.org/I130701444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100403882","display_name":"Grace Kim","orcid":"https://orcid.org/0000-0003-4603-9448"},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Grace C. Kim","raw_affiliation_strings":["Georgia Tech"],"affiliations":[{"raw_affiliation_string":"Georgia Tech","institution_ids":["https://openalex.org/I130701444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006255387","display_name":"Shengyun Peng","orcid":"https://orcid.org/0000-0003-3063-2052"},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"ShengYun Peng","raw_affiliation_strings":["Georgia Tech"],"affiliations":[{"raw_affiliation_string":"Georgia Tech","institution_ids":["https://openalex.org/I130701444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120308958","display_name":"Mansi Phute","orcid":null},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mansi Phute","raw_affiliation_strings":["Georgia Tech"],"affiliations":[{"raw_affiliation_string":"Georgia Tech","institution_ids":["https://openalex.org/I130701444"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5020153026","display_name":"Duen Horng Chau","orcid":"https://orcid.org/0000-0001-9824-3323"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Duen Horng Chau","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5101442992"],"corresponding_institution_ids":["https://openalex.org/I130701444"],"apc_list":null,"apc_paid":null,"fwci":1.3988,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.85681156,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"21514","last_page":"21545"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T13295","display_name":"Safety Systems Engineering in Autonomy","score":0.7422999739646912,"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/T13295","display_name":"Safety Systems Engineering in Autonomy","score":0.7422999739646912,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.014499999582767487,"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/T10142","display_name":"Formal Methods in Verification","score":0.013700000010430813,"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/interpretation","display_name":"Interpretation (philosophy)","score":0.7014999985694885},{"id":"https://openalex.org/keywords/field","display_name":"Field (mathematics)","score":0.3068000078201294},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.2578999996185303},{"id":"https://openalex.org/keywords/data-collection","display_name":"Data collection","score":0.25290000438690186}],"concepts":[{"id":"https://openalex.org/C527412718","wikidata":"https://www.wikidata.org/wiki/Q855395","display_name":"Interpretation (philosophy)","level":2,"score":0.7014999985694885},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.489300012588501},{"id":"https://openalex.org/C539667460","wikidata":"https://www.wikidata.org/wiki/Q2414942","display_name":"Management science","level":1,"score":0.3472000062465668},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.3188999891281128},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.3068000078201294},{"id":"https://openalex.org/C55587333","wikidata":"https://www.wikidata.org/wiki/Q1133029","display_name":"Engineering ethics","level":1,"score":0.2879999876022339},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.267300009727478},{"id":"https://openalex.org/C111472728","wikidata":"https://www.wikidata.org/wiki/Q9471","display_name":"Epistemology","level":1,"score":0.2603999972343445},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.2578999996185303},{"id":"https://openalex.org/C133462117","wikidata":"https://www.wikidata.org/wiki/Q4929239","display_name":"Data collection","level":2,"score":0.25290000438690186}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.emnlp-main.1091","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.1091","pdf_url":"https://aclanthology.org/2025.emnlp-main.1091.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 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.emnlp-main.1091","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.1091","pdf_url":"https://aclanthology.org/2025.emnlp-main.1091.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 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416035991.pdf","grobid_xml":"https://content.openalex.org/works/W4416035991.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"As":[0],"large":[1],"language":[2],"models":[3],"(LLMs)":[4],"see":[5],"wider":[6],"real-world":[7],"use,":[8],"understanding":[9],"and":[10,24,60,84,91],"mitigating":[11],"their":[12,78],"unsafe":[13,22],"behaviors":[14],"is":[15],"critical.Interpretation":[16],"techniques":[17],"can":[18],"reveal":[19],"causes":[20],"of":[21],"outputs":[23],"guide":[25],"safety,":[26],"but":[27],"such":[28],"connections":[29],"with":[30,81],"safety":[31,56],"are":[32],"often":[33],"overlooked":[34],"in":[35],"prior":[36],"surveys.We":[37],"present":[38],"the":[39,55,61],"first":[40],"survey":[41,88],"that":[42,50,63],"bridges":[43],"this":[44],"gap,":[45],"introducing":[46],"a":[47],"unified":[48],"framework":[49],"connects":[51],"safety-focused":[52],"interpretation":[53],"methods,":[54],"enhancements":[57],"they":[58],"inform,":[59],"tools":[62],"operationalize":[64],"them.Our":[65],"novel":[66],"taxonomy,":[67],"organized":[68],"by":[69],"LLM":[70],"workflow":[71],"stages,":[72],"summarizes":[73],"nearly":[74],"70":[75],"works":[76],"at":[77],"intersections.We":[79],"conclude":[80],"open":[82],"challenges":[83],"future":[85],"directions.This":[86],"timely":[87],"helps":[89],"researchers":[90],"practitioners":[92],"navigate":[93],"key":[94],"advancements":[95],"for":[96],"safer,":[97],"more":[98],"interpretable":[99],"LLMs.":[100]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-03-14T06:41:57.775601","created_date":"2025-11-08T00:00:00"}