{"id":"https://openalex.org/W4415316983","doi":"https://doi.org/10.18653/v1/2026.findings-acl.62","title":"Differentially Private Synthetic Text Generation for Retrieval-Augmented Generation (RAG)","display_name":"Differentially Private Synthetic Text Generation for Retrieval-Augmented Generation (RAG)","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W4415316983","doi":"https://doi.org/10.18653/v1/2026.findings-acl.62"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.62","doi":"https://doi.org/10.18653/v1/2026.findings-acl.62","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.62","pdf_url":"https://aclanthology.org/2026.findings-acl.62.pdf","source":{"id":"https://openalex.org/S4363605144","display_name":"Findings of the Association for Computational Linguistics: ACL (Findings of ACL)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"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: ACL 2026","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://aclanthology.org/2026.findings-acl.62.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5035060444","display_name":"Junki Mori","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I118347220","display_name":"NEC (Japan)","ror":"https://ror.org/04jndar25","country_code":"JP","type":"company","lineage":["https://openalex.org/I118347220"]},{"id":"https://openalex.org/I4210107353","display_name":"NEC (United States)","ror":"https://ror.org/01v791m31","country_code":"US","type":"company","lineage":["https://openalex.org/I118347220","https://openalex.org/I4210107353"]}],"countries":["JP","US"],"is_corresponding":false,"raw_author_name":"Junki Mori","raw_affiliation_strings":["NEC Corporation ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NEC Corporation ,","institution_ids":["https://openalex.org/I118347220","https://openalex.org/I4210107353"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055148101","display_name":"Kazuya Kakizaki","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I118347220","display_name":"NEC (Japan)","ror":"https://ror.org/04jndar25","country_code":"JP","type":"company","lineage":["https://openalex.org/I118347220"]},{"id":"https://openalex.org/I4210107353","display_name":"NEC (United States)","ror":"https://ror.org/01v791m31","country_code":"US","type":"company","lineage":["https://openalex.org/I118347220","https://openalex.org/I4210107353"]}],"countries":["JP","US"],"is_corresponding":false,"raw_author_name":"Kazuya Kakizaki","raw_affiliation_strings":["NEC Corporation ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NEC Corporation ,","institution_ids":["https://openalex.org/I118347220","https://openalex.org/I4210107353"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071714722","display_name":"Taiki Miyagawa","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I118347220","display_name":"NEC (Japan)","ror":"https://ror.org/04jndar25","country_code":"JP","type":"company","lineage":["https://openalex.org/I118347220"]},{"id":"https://openalex.org/I4210107353","display_name":"NEC (United States)","ror":"https://ror.org/01v791m31","country_code":"US","type":"company","lineage":["https://openalex.org/I118347220","https://openalex.org/I4210107353"]}],"countries":["JP","US"],"is_corresponding":false,"raw_author_name":"Taiki Miyagawa","raw_affiliation_strings":["NEC Corporation ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NEC Corporation ,","institution_ids":["https://openalex.org/I118347220","https://openalex.org/I4210107353"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5022139141","display_name":"Jun Sakuma","orcid":"https://orcid.org/0000-0001-5015-3812","observed_orcids":["https://orcid.org/0000-0001-5015-3812"]},"institutions":[{"id":"https://openalex.org/I4210126580","display_name":"RIKEN Center for Advanced Intelligence Project","ror":"https://ror.org/03ckxwf91","country_code":"JP","type":"facility","lineage":["https://openalex.org/I4210110652","https://openalex.org/I4210126580"]},{"id":"https://openalex.org/I4400009020","display_name":"Institute of Science Tokyo","ror":"https://ror.org/05dqf9946","country_code":"JP","type":"education","lineage":["https://openalex.org/I4400009020"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Jun Sakuma","raw_affiliation_strings":["Institute of Science Tokyo ,","RIKEN Center for Advanced Intelligence Project"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Science Tokyo ,","institution_ids":["https://openalex.org/I4400009020"]},{"raw_affiliation_string":"RIKEN Center for Advanced Intelligence Project","institution_ids":["https://openalex.org/I4210126580"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":8.5469,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.92783053,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":93,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"1216","last_page":"1235"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9994999766349792,"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":0.9994999766349792,"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/T10028","display_name":"Topic Modeling","score":9.999999747378752e-05,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.0,"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/retrieval-augmented-generation","display_name":"retrieval-augmented generation","score":0.8980000019073486},{"id":"https://openalex.org/keywords/differential-privacy","display_name":"differential privacy","score":0.9769999980926514},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.7200000286102295},{"id":"https://openalex.org/keywords/noise-injection","display_name":"noise injection","score":0.5170000195503235},{"id":"https://openalex.org/keywords/privacy-budget","display_name":"privacy budget","score":0.5109999775886536}],"concepts":[{"id":"https://openalex.org/C23130292","wikidata":"https://www.wikidata.org/wiki/Q5275358","display_name":"Differential privacy","level":2,"score":0.7558000087738037},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6836000084877014},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6381999850273132},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.5794000029563904},{"id":"https://openalex.org/C99221444","wikidata":"https://www.wikidata.org/wiki/Q1532069","display_name":"Private information retrieval","level":2,"score":0.5446000099182129},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.5008999705314636},{"id":"https://openalex.org/C137822555","wikidata":"https://www.wikidata.org/wiki/Q2587068","display_name":"Information sensitivity","level":2,"score":0.43849998712539673},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.43309998512268066},{"id":"https://openalex.org/C2780909371","wikidata":"https://www.wikidata.org/wiki/Q4801092","display_name":"Artificial noise","level":4,"score":0.41110000014305115},{"id":"https://openalex.org/C123201435","wikidata":"https://www.wikidata.org/wiki/Q456632","display_name":"Information privacy","level":2,"score":0.4041999876499176},{"id":"https://openalex.org/C2910417920","wikidata":"https://www.wikidata.org/wiki/Q4116434","display_name":"Patient privacy","level":3,"score":0.37299999594688416},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.37279999256134033},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3472000062465668},{"id":"https://openalex.org/C108827166","wikidata":"https://www.wikidata.org/wiki/Q175975","display_name":"Internet privacy","level":1,"score":0.3449000120162964},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3425999879837036},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.32670000195503235},{"id":"https://openalex.org/C2985684807","wikidata":"https://www.wikidata.org/wiki/Q1513879","display_name":"Text generation","level":2,"score":0.31310001015663147},{"id":"https://openalex.org/C2781170535","wikidata":"https://www.wikidata.org/wiki/Q30587856","display_name":"Noisy data","level":2,"score":0.3075999915599823},{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.2994999885559082},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.29429998993873596},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.29260000586509705},{"id":"https://openalex.org/C75684735","wikidata":"https://www.wikidata.org/wiki/Q858810","display_name":"Big data","level":2,"score":0.28859999775886536},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.28600001335144043},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.2533000111579895}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.62","doi":"https://doi.org/10.18653/v1/2026.findings-acl.62","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.62","pdf_url":"https://aclanthology.org/2026.findings-acl.62.pdf","source":{"id":"https://openalex.org/S4363605144","display_name":"Findings of the Association for Computational Linguistics: ACL (Findings of ACL)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"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: ACL 2026","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2510.06719","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2510.06719","pdf_url":"https://arxiv.org/pdf/2510.06719","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2510.06719","doi":"https://doi.org/10.48550/arxiv.2510.06719","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2510.06719","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,"listed_in":[],"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":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.62","doi":"https://doi.org/10.18653/v1/2026.findings-acl.62","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.62","pdf_url":"https://aclanthology.org/2026.findings-acl.62.pdf","source":{"id":"https://openalex.org/S4363605144","display_name":"Findings of the Association for Computational Linguistics: ACL (Findings of ACL)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"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: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4415316983.pdf","grobid_xml":"https://content.openalex.org/works/W4415316983.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7202048339","https://openalex.org/W4405173318","https://openalex.org/W7129726123","https://openalex.org/W4399986892","https://openalex.org/W4412082268","https://openalex.org/W4408977202","https://openalex.org/W7166886659","https://openalex.org/W4416189337","https://openalex.org/W7169610995","https://openalex.org/W4404781492"],"abstract_inverted_index":{"Retrieval-Augmented":[0],"Generation":[1],"(RAG)":[2],"enhances":[3],"large":[4],"language":[5],"models":[6],"(LLMs)":[7],"by":[8,21],"grounding":[9],"them":[10],"in":[11,16,104],"external":[12],"knowledge.However,":[13],"its":[14],"application":[15],"sensitive":[17],"domains":[18],"is":[19],"limited":[20],"privacy":[22,32,43,80,124],"risks.Existing":[23],"private":[24,59,91,117],"RAG":[25,61,87,118],"methods":[26],"typically":[27],"rely":[28],"on":[29],"query-time":[30],"differential":[31],"(DP),":[33],"which":[34,93],"requires":[35],"repeated":[36,75],"noise":[37,76],"injection":[38,77],"and":[39,78],"leads":[40],"to":[41,56,96,114],"accumulated":[42],"loss.To":[44],"address":[45],"this":[46],"issue,":[47],"we":[48],"propose":[49],"DP-SynRAG,":[50],"a":[51,105,122,127],"framework":[52],"that":[53,99,109],"uses":[54],"LLMs":[55,95],"generate":[57,97],"differentially":[58],"synthetic":[60,66],"databases.Unlike":[62],"prior":[63],"methods,":[64],"the":[65,115],"text":[67,98],"can":[68],"be":[69],"reused":[70],"once":[71],"created,":[72],"thereby":[73],"avoiding":[74],"additional":[79],"costs.To":[81],"preserve":[82],"essential":[83],"information":[84],"for":[85,130],"downstream":[86],"tasks,":[88],"DP-SynRAG":[89,110],"extends":[90],"prediction,":[92],"instructs":[94],"mimics":[100],"subsampled":[101],"database":[102],"records":[103],"DP":[106],"manner.Experiments":[107],"show":[108],"achieves":[111],"superior":[112],"performance":[113],"state-of-the-art":[116],"systems":[119],"while":[120],"maintaining":[121],"fixed":[123],"budget,":[125],"offering":[126],"scalable":[128],"solution":[129],"privacy-preserving":[131],"RAG.\u2026":[132]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-10-26T11:51:55.731654","created_date":"2025-10-18T00:00:00"}