{"id":"https://openalex.org/W4416139154","doi":"https://doi.org/10.18653/v1/2026.acl-long.756","title":"Why LLM Safety Guardrails Collapse After Fine-tuning: A Similarity Analysis Between Alignment and Fine-tuning Datasets","display_name":"Why LLM Safety Guardrails Collapse After Fine-tuning: A Similarity Analysis Between Alignment and Fine-tuning Datasets","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W4416139154","doi":"https://doi.org/10.18653/v1/2026.acl-long.756"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2026.acl-long.756","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.756","pdf_url":"https://aclanthology.org/2026.acl-long.756.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 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.acl-long.756.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5120395291","display_name":"Lei Hsiung","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lei Hsiung","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120443247","display_name":"Tianyu Pang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tianyu Pang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Yung-Chen Tang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yung-Chen Tang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103466920","display_name":"Linyue Song","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Linyue Song","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062800747","display_name":"Tsung-Yi Ho","orcid":"https://orcid.org/0000-0001-7348-5625"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tsung-Yi Ho","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119947959","display_name":"Pin-Yu Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pin-Yu Chen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"Yaoqing Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yaoqing Yang","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":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.02112713,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"16601","last_page":"16619"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.6500999927520752,"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":0.6500999927520752,"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/T10734","display_name":"Information and Cyber Security","score":0.05429999902844429,"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/T10028","display_name":"Topic Modeling","score":0.053599998354911804,"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/upstream","display_name":"Upstream (networking)","score":0.6740000247955322},{"id":"https://openalex.org/keywords/downstream","display_name":"Downstream (manufacturing)","score":0.6151000261306763},{"id":"https://openalex.org/keywords/similarity","display_name":"Similarity (geometry)","score":0.5806000232696533},{"id":"https://openalex.org/keywords/vulnerability","display_name":"Vulnerability (computing)","score":0.545199990272522},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.48420000076293945}],"concepts":[{"id":"https://openalex.org/C191172861","wikidata":"https://www.wikidata.org/wiki/Q7899321","display_name":"Upstream (networking)","level":2,"score":0.6740000247955322},{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.6151000261306763},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5978999733924866},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.5806000232696533},{"id":"https://openalex.org/C95713431","wikidata":"https://www.wikidata.org/wiki/Q631425","display_name":"Vulnerability (computing)","level":2,"score":0.545199990272522},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.48420000076293945},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.44780001044273376},{"id":"https://openalex.org/C158980903","wikidata":"https://www.wikidata.org/wiki/Q7899329","display_name":"Upstream and downstream (DNA)","level":3,"score":0.3695000112056732},{"id":"https://openalex.org/C167063184","wikidata":"https://www.wikidata.org/wiki/Q1400839","display_name":"Vulnerability assessment","level":3,"score":0.34929999709129333},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.31290000677108765},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.3089999854564667},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.26739999651908875}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.756","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.756","pdf_url":"https://aclanthology.org/2026.acl-long.756.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 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2506.05346","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2506.05346","pdf_url":"https://arxiv.org/pdf/2506.05346","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2506.05346","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2506.05346","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.acl-long.756","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.756","pdf_url":"https://aclanthology.org/2026.acl-long.756.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 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320306084","display_name":"U.S. Department of Energy","ror":"https://ror.org/01bj3aw27"},{"id":"https://openalex.org/F4320310400","display_name":"Dartmouth College","ror":"https://ror.org/049s0rh22"},{"id":"https://openalex.org/F4320332180","display_name":"Defense Advanced Research Projects Agency","ror":"https://ror.org/02caytj08"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416139154.pdf","grobid_xml":"https://content.openalex.org/works/W4416139154.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Lei":[0],"Hsiung,":[1],"Tianyu":[2],"Pang,":[3],"Yung-Chen":[4],"Tang,":[5],"Linyue":[6],"Song,":[7],"Tsung-Yi":[8],"Ho,":[9],"Pin-Yu":[10],"Chen,":[11],"Yaoqing":[12],"Yang.":[13],"Proceedings":[14],"of":[15,20],"the":[16,21],"64th":[17],"Annual":[18],"Meeting":[19],"Association":[22],"for":[23],"Computational":[24],"Linguistics":[25],"(Volume":[26],"1:":[27],"Long":[28],"Papers).":[29],"2026.":[30]},"counts_by_year":[],"updated_date":"2026-08-05T07:39:15.569665","created_date":"2025-10-10T00:00:00"}