{"id":"https://openalex.org/W4412945490","doi":"https://doi.org/10.18653/v1/2025.acl-long.475","title":"PIG: Privacy Jailbreak Attack on LLMs via Gradient-based Iterative In-Context Optimization","display_name":"PIG: Privacy Jailbreak Attack on LLMs via Gradient-based Iterative In-Context Optimization","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412945490","doi":"https://doi.org/10.18653/v1/2025.acl-long.475"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.475","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.475","pdf_url":"https://aclanthology.org/2025.acl-long.475.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 63rd 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/2025.acl-long.475.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5003880609","display_name":"Yidan Wang","orcid":"https://orcid.org/0000-0002-0741-6506"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yidan Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100595530","display_name":"Yanan Cao","orcid":"https://orcid.org/0000-0001-8030-9507"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yanan Cao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101219998","display_name":"Yubing Ren","orcid":"https://orcid.org/0000-0002-0815-3998"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yubing Ren","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100386895","display_name":"Fang Fang","orcid":"https://orcid.org/0000-0003-3784-3696"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fang Fang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089094818","display_name":"Zheng Lin","orcid":"https://orcid.org/0000-0002-4463-5652"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng Lin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5091368858","display_name":"Binxing Fang","orcid":"https://orcid.org/0000-0003-0305-2132"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Binxing Fang","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":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"9645","last_page":"9660"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9830999970436096,"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":0.9830999970436096,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9664999842643738,"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/T11241","display_name":"Advanced Malware Detection Techniques","score":0.9366999864578247,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/context","display_name":"Context (archaeology)","score":0.6092088222503662},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6031612157821655},{"id":"https://openalex.org/keywords/computer-security","display_name":"Computer security","score":0.498842716217041},{"id":"https://openalex.org/keywords/internet-privacy","display_name":"Internet privacy","score":0.4672078490257263}],"concepts":[{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6092088222503662},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6031612157821655},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.498842716217041},{"id":"https://openalex.org/C108827166","wikidata":"https://www.wikidata.org/wiki/Q175975","display_name":"Internet privacy","level":1,"score":0.4672078490257263},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.475","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.475","pdf_url":"https://aclanthology.org/2025.acl-long.475.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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.acl-long.475","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.475","pdf_url":"https://aclanthology.org/2025.acl-long.475.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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions","score":0.5}],"awards":[{"id":"https://openalex.org/G5077752208","display_name":null,"funder_award_id":"U2336202","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412945490.pdf","grobid_xml":"https://content.openalex.org/works/W4412945490.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":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"excel":[4],"in":[5,18,51,65,75,103,159],"various":[6],"domains":[7],"but":[8,48],"pose":[9],"inherent":[10],"privacy":[11,16,52,70,104,112,157],"risks.Existing":[12],"methods":[13,131,148],"to":[14,27,44,109,122],"evaluate":[15,126],"leakage":[17,71],"LLMs":[19,142],"often":[20],"use":[21],"memorized":[22],"prefixes":[23],"or":[24],"simple":[25],"instructions":[26],"extract":[28],"data,":[29],"both":[30],"of":[31,62,92],"which":[32],"well-alignment":[33],"models":[34],"can":[35],"easily":[36],"block.Meanwhile,":[37],"Jailbreak":[38],"attacks":[39,64,74],"bypass":[40],"LLM":[41],"safety":[42],"mechanisms":[43],"generate":[45],"harmful":[46],"content,":[47],"their":[49,101],"role":[50],"scenarios":[53],"remains":[54],"underexplored.In":[55],"this":[56],"paper,":[57],"we":[58,77],"examine":[59],"the":[60,90,162],"effectiveness":[61],"jailbreak":[63,73,94,130],"extracting":[66],"sensitive":[67],"information,":[68],"bridging":[69],"and":[72,88,100,114,128,139,149],"LLMs.Moreover,":[76],"propose":[78],"PIG,":[79],"a":[80,111],"novel":[81],"framework":[82],"targeting":[83],"Personally":[84],"Identifiable":[85],"Information":[86],"(PII)":[87],"addressing":[89],"limitations":[91],"current":[93],"methods.Specifically,":[95],"PIG":[96,127,145],"identifies":[97],"PII":[98],"entities":[99],"types":[102],"queries,":[105],"uses":[106],"in-context":[107],"learning":[108],"build":[110],"context,":[113],"iteratively":[115],"updates":[116],"it":[117],"with":[118],"three":[119],"gradient-based":[120],"strategies":[121],"elicit":[123],"target":[124],"PII.We":[125],"existing":[129],"using":[132],"two":[133,140],"privacy-related":[134],"datasets.Experiments":[135],"on":[136],"four":[137],"white-box":[138],"blackbox":[141],"show":[143],"that":[144],"outperforms":[146],"baseline":[147],"achieves":[150],"state-of-the-art":[151],"(SoTA)":[152],"results.The":[153],"results":[154],"underscore":[155],"significant":[156],"risks":[158],"LLMs,":[160],"emphasizing":[161],"need":[163],"for":[164],"stronger":[165],"safeguards.Our":[166],"code":[167],"is":[168],"availble":[169],"at":[170],"https://github.com/redwyd/PrivacyJailbreak.":[171]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}