{"id":"https://openalex.org/W7154945076","doi":"https://doi.org/10.48550/arxiv.2604.15660","title":"DPDSyn: Improving Differentially Private Dataset Synthesis for Model Training by Downstream Task Guidance","display_name":"DPDSyn: Improving Differentially Private Dataset Synthesis for Model Training by Downstream Task Guidance","publication_year":2026,"publication_date":"2026-04-17","ids":{"openalex":"https://openalex.org/W7154945076","doi":"https://doi.org/10.48550/arxiv.2604.15660"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.15660","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.15660","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.15660","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5049685388","display_name":"Mingxuan Jia","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jia, Mingxuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134036177","display_name":"Wen Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Wen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134018434","display_name":"Weixin Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Weixin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134030776","display_name":"Xingyi Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Xingyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134050610","display_name":"Jian Peng","orcid":"https://orcid.org/0009-0009-2140-4518"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peng, Jian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5134029226","display_name":"Zhishuo Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Zhishuo","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":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9689000248908997,"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":0.9689000248908997,"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":0.0035000001080334187,"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/T10237","display_name":"Cryptography and Data Security","score":0.0031999999191612005,"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/differential-privacy","display_name":"Differential privacy","score":0.77920001745224},{"id":"https://openalex.org/keywords/downstream","display_name":"Downstream (manufacturing)","score":0.7146000266075134},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.7073000073432922},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.6310999989509583},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.600600004196167},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.5415999889373779},{"id":"https://openalex.org/keywords/synthetic-data","display_name":"Synthetic data","score":0.49959999322891235},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.41499999165534973}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7817999720573425},{"id":"https://openalex.org/C23130292","wikidata":"https://www.wikidata.org/wiki/Q5275358","display_name":"Differential privacy","level":2,"score":0.77920001745224},{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.7146000266075134},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.7073000073432922},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6310999989509583},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.600600004196167},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5679000020027161},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5562000274658203},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.5512999892234802},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.5415999889373779},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.49959999322891235},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.41499999165534973},{"id":"https://openalex.org/C99221444","wikidata":"https://www.wikidata.org/wiki/Q1532069","display_name":"Private information retrieval","level":2,"score":0.40639999508857727},{"id":"https://openalex.org/C93226319","wikidata":"https://www.wikidata.org/wiki/Q193137","display_name":"Differential (mechanical device)","level":2,"score":0.359499990940094},{"id":"https://openalex.org/C35639132","wikidata":"https://www.wikidata.org/wiki/Q7452468","display_name":"Sequence labeling","level":3,"score":0.35749998688697815},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.35580000281333923},{"id":"https://openalex.org/C191172861","wikidata":"https://www.wikidata.org/wiki/Q7899321","display_name":"Upstream (networking)","level":2,"score":0.2939000129699707},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.26919999718666077},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.2605000138282776},{"id":"https://openalex.org/C2780440489","wikidata":"https://www.wikidata.org/wiki/Q5227278","display_name":"Data-driven","level":2,"score":0.25679999589920044},{"id":"https://openalex.org/C110121322","wikidata":"https://www.wikidata.org/wiki/Q865811","display_name":"Distribution (mathematics)","level":2,"score":0.2551000118255615}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.15660","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.15660","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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:10.48550/arxiv.2604.15660","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.15660","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"How":[0],"to":[1,37,66,90,110,129,151,172,175],"synthesize":[2,50,111,173],"a":[3,14,51,92,203],"dataset":[4,44,52,86,104],"while":[5,183],"achieving":[6],"differential":[7,123],"privacy":[8,124],"for":[9,97,163],"AI":[10,95,120,147,158,170],"model":[11,96,109,121,132,148,159,171],"training":[12],"is":[13,64,149],"meaningful":[15],"but":[16],"challenging":[17],"problem.":[18],"To":[19,82],"address":[20],"this":[21],"problem,":[22],"state-of-the-art":[23,200],"methods":[24],"first":[25],"select":[26,67],"original":[27,42,102,139],"private":[28,43,85,94,103,136,140],"dataset's":[29],"multiple":[30],"low-dimensional":[31,56,69],"distributions":[32,57],"that":[33,193,219],"have":[34],"the":[35,39,59,75,101,107,116,119,131,143,146,153,157,169,177],"potential":[36],"approximate":[38],"distribution":[40],"of":[41,78,138,179,206],"with":[45,202],"high":[46],"precision,":[47],"and":[48,105,210],"then":[49],"obeying":[53],"all":[54],"selected":[55],"as":[58],"synthetic":[60,80],"dataset.":[61,81,141],"However,":[62],"it":[63],"difficult":[65],"suitable":[68],"distributions,":[70],"which":[71],"in":[72,208,212],"turn":[73],"degrades":[74],"data":[76,181,226],"utility":[77,182],"resulting":[79],"improve":[83],"differentially":[84,93],"synthesis,":[87],"we":[88],"propose":[89],"train":[91],"downstream":[98,154,165],"tasks":[99],"on":[100,115,188],"utilize":[106,168],"trained":[108,150],"datasets.":[112],"In":[113],"particular,":[114],"one":[117],"hand,":[118,145],"satisfies":[122],"so":[125,156],"no":[126],"matter":[127],"how":[128],"use":[130],"does":[133],"not":[134],"disclose":[135],"information":[137,162],"On":[142],"other":[144],"complete":[152],"task":[155],"preserves":[160],"critical":[161],"completing":[164],"tasks.":[166],"We":[167],"datasets":[174,191],"achieve":[176],"goal":[178],"improving":[180],"preserving":[184],"privacy.":[185],"Empirical":[186],"evaluations":[187],"four":[189],"benchmark":[190],"demonstrate":[192],"our":[194],"proposed":[195],"DPDSyn":[196,220],"consistently":[197],"outperforms":[198],"eight":[199],"baselines":[201],"maximum":[204],"improvement":[205],"2.40x":[207],"accuracy":[209],"333.73x":[211],"synthesis":[213],"efficiency.":[214],"Further":[215],"experiments":[216],"also":[217],"validate":[218],"has":[221],"strong":[222],"scalability":[223],"across":[224],"varying":[225],"scales.":[227]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-21T00:00:00"}