{"id":"https://openalex.org/W4416034702","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.778","title":"Harry Potter is Still Here! Probing Knowledge Leakage in Targeted Unlearned Large Language Models","display_name":"Harry Potter is Still Here! Probing Knowledge Leakage in Targeted Unlearned Large Language Models","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416034702","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.778"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.778","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.778","pdf_url":"https://aclanthology.org/2025.findings-emnlp.778.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.778.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5120308661","display_name":"Bang Trinh Tran To","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bang Trinh Tran To","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5024109615","display_name":"Th\u00e1i Ho\u00e0ng L\u00ea","orcid":"https://orcid.org/0000-0002-6801-1924"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Thai Le","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":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"14427","last_page":"14439"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.3199999928474426,"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":0.3199999928474426,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.07999999821186066,"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.07769999653100967,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/harry-potter","display_name":"Harry potter","score":0.6007999777793884},{"id":"https://openalex.org/keywords/leakage","display_name":"Leakage (economics)","score":0.45669999718666077},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.2793000042438507},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.2605000138282776}],"concepts":[{"id":"https://openalex.org/C2992468625","wikidata":"https://www.wikidata.org/wiki/Q8337","display_name":"Harry potter","level":2,"score":0.6007999777793884},{"id":"https://openalex.org/C2777042071","wikidata":"https://www.wikidata.org/wiki/Q6509304","display_name":"Leakage (economics)","level":2,"score":0.45669999718666077},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.4514999985694885},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.4458000063896179},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.4424000084400177},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.36320000886917114},{"id":"https://openalex.org/C46312422","wikidata":"https://www.wikidata.org/wiki/Q11024","display_name":"Communication","level":1,"score":0.34220001101493835},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3133000135421753},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2793000042438507},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.2711000144481659},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.2605000138282776}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.778","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.778","pdf_url":"https://aclanthology.org/2025.findings-emnlp.778.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.778","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.778","pdf_url":"https://aclanthology.org/2025.findings-emnlp.778.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":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416034702.pdf","grobid_xml":"https://content.openalex.org/works/W4416034702.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"This":[0],"work":[1],"presents":[2],"LURK":[3,74],"(Latent":[4],"UnleaRned":[5],"Knowledge),":[6],"a":[7,38,76],"novel":[8],"framework":[9],"that":[10,46],"probes":[11],"for":[12,42,82],"undesired":[13],"knowledge":[14,32,70],"retention":[15],"in":[16],"unlearned":[17,51],"LLMs":[18],"through":[19,71],"adversarial":[20,25,58],"suffix":[21],"prompting.LURK":[22],"automatically":[23],"generates":[24],"prompt":[26],"suffixes":[27],"designed":[28],"to":[29],"elicit":[30],"residual":[31],"about":[33],"the":[34,84],"Harry":[35],"Potter":[36],"domain,":[37],"commonly":[39],"used":[40],"benchmark":[41],"unlearning.Our":[43],"experiments":[44],"reveal":[45],"even":[47],"models":[48],"deemed":[49],"successfully":[50],"can":[52],"leak":[53],"idiosyncratic":[54],"information":[55],"under":[56],"targeted":[57],"conditions,":[59],"highlighting":[60],"critical":[61],"limitations":[62],"of":[63,86],"current":[64],"unlearning":[65,87],"evaluation":[66],"standards.By":[67],"uncovering":[68],"implicit":[69],"indirect":[72],"probing,":[73],"offers":[75],"more":[77],"rigorous":[78],"and":[79,89],"diagnostic":[80],"tool":[81],"assessing":[83],"robustness":[85],"algorithms.Code":[88],"data":[90],"will":[91],"be":[92],"available":[93],"at":[94],"https://github.com/Rachel1809/LURK.":[95]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-11-08T00:00:00"}