{"id":"https://openalex.org/W7166869521","doi":"https://doi.org/10.18653/v1/2026.acl-long.400","title":"Privacy Collapse: Benign Fine-Tuning Can Break Contextual Privacy in Language Models","display_name":"Privacy Collapse: Benign Fine-Tuning Can Break Contextual Privacy in Language Models","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166869521","doi":"https://doi.org/10.18653/v1/2026.acl-long.400"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2026.acl-long.400","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.400","pdf_url":"https://aclanthology.org/2026.acl-long.400.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":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.acl-long.400.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5051727084","display_name":"Anmol Goel","orcid":"https://orcid.org/0009-0009-4807-2402"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Anmol Goel","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005232344","display_name":"Cornelius Emde","orcid":"https://orcid.org/0000-0003-0348-6109"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cornelius Emde","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139735964","display_name":"Seong Joon Oh","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Seong Joon Oh","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139770105","display_name":"Sangdoo Yun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sangdoo Yun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5058150999","display_name":"Martin Gubri","orcid":"https://orcid.org/0000-0001-6744-6662"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Martin Gubri","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":0.8582828,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"8870","last_page":"8892"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.27970001101493835,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.27970001101493835,"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/T12380","display_name":"Authorship Attribution and Profiling","score":0.044199999421834946,"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/T14347","display_name":"Big Data and Digital Economy","score":0.03319999948143959,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/patient-privacy","display_name":"Patient privacy","score":0.398499995470047},{"id":"https://openalex.org/keywords/information-privacy","display_name":"Information privacy","score":0.38929998874664307},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.37059998512268066},{"id":"https://openalex.org/keywords/privacy-software","display_name":"Privacy software","score":0.35830000042915344},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.33090001344680786},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.2906999886035919}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6144000291824341},{"id":"https://openalex.org/C108827166","wikidata":"https://www.wikidata.org/wiki/Q175975","display_name":"Internet privacy","level":1,"score":0.39980000257492065},{"id":"https://openalex.org/C2910417920","wikidata":"https://www.wikidata.org/wiki/Q4116434","display_name":"Patient privacy","level":3,"score":0.398499995470047},{"id":"https://openalex.org/C123201435","wikidata":"https://www.wikidata.org/wiki/Q456632","display_name":"Information privacy","level":2,"score":0.38929998874664307},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.37059998512268066},{"id":"https://openalex.org/C509729295","wikidata":"https://www.wikidata.org/wiki/Q7246032","display_name":"Privacy software","level":3,"score":0.35830000042915344},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.33090001344680786},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3061000108718872},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.30250000953674316},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.2906999886035919},{"id":"https://openalex.org/C102938260","wikidata":"https://www.wikidata.org/wiki/Q1999831","display_name":"Privacy policy","level":3,"score":0.28060001134872437},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.27950000762939453},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.27950000762939453},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.26759999990463257},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.26499998569488525},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.2623000144958496}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.400","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.400","pdf_url":"https://aclanthology.org/2026.acl-long.400.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:tubiblio.ulb.tu-darmstadt.de:161132","is_oa":false,"landing_page_url":"http://tubiblio.ulb.tu-darmstadt.de/view/person/Goel=3AAnmol=3A=3A.html>","pdf_url":null,"source":{"id":"https://openalex.org/S4377196390","display_name":"TUbilio (Technical University of Darmstadt)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I31512782","host_organization_name":"Technische Universit\u00e4t Darmstadt","host_organization_lineage":["https://openalex.org/I31512782"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"Konferenzver\u00f6ffentlichung"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.acl-long.400","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.400","pdf_url":"https://aclanthology.org/2026.acl-long.400.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":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166869521.pdf","grobid_xml":"https://content.openalex.org/works/W7166869521.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0],"identify":[1],"a":[2,73,139],"novel":[3],"phenomenon":[4],"in":[5,23,142,146],"language":[6],"models:":[7],"benign":[8],"fine-tuning":[9,106],"of":[10,95,151],"frontier":[11],"models":[12,50,77,100],"can":[13,26],"lead":[14],"to":[15,35,54,128,131],"privacy":[16,58,90,96,123],"collapse.We":[17],"find":[18],"that":[19,122],"diverse,":[20],"subtle":[21],"patterns":[22],"training":[24],"data":[25],"degrade":[27],"contextual":[28,57],"privacy,":[29],"including":[30],"optimisation":[31],"for":[32,148],"helpfulness,":[33],"exposure":[34],"user":[36],"information,":[37],"emotional":[38],"and":[39,42,65,84,102,109,112,117],"subjective":[40],"dialogue,":[41],"debugging":[43],"code":[44],"printing":[45],"internal":[46],"variables,":[47],"among":[48],"others.Fine-tuned":[49],"lose":[51],"their":[52],"ability":[53],"reason":[55],"about":[56],"norms,":[59],"share":[60],"information":[61],"inappropriately":[62],"with":[63],"tools,":[64],"violate":[66],"memory":[67],"boundaries":[68],"across":[69,98],"contexts.Privacy":[70],"collapse":[71,97],"is":[72],"\"silent":[74],"failure\"":[75],"because":[76],"maintain":[78],"high":[79],"performance":[80],"on":[81],"standard":[82],"safety":[83,144],"utility":[85],"benchmarks":[86],"whilst":[87],"exhibiting":[88],"severe":[89],"vulnerabilities.Our":[91],"experiments":[92],"show":[93],"evidence":[94],"six":[99],"(closed":[101],"open":[103],"weight),":[104],"five":[105],"datasets":[107],"(real-world":[108],"controlled":[110],"data),":[111],"two":[113],"task":[114],"categories":[115],"(agentic":[116],"memory-based).Our":[118],"mechanistic":[119],"analysis":[120],"reveals":[121],"representations":[124],"are":[125,135],"uniquely":[126],"fragile":[127],"fine-tuning,":[129],"compared":[130],"task-relevant":[132],"features":[133],"which":[134],"preserved.Our":[136],"results":[137],"reveal":[138],"critical":[140],"gap":[141],"current":[143],"evaluations,":[145],"particular":[147],"the":[149],"deployment":[150],"specialised":[152],"agents.":[153]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}