Privacy-Enhanced Database Synthesis for Benchmark Publishing

Summary: PrivBench uses sum-product networks for differentially private synthesis of multi-relational benchmark databases, preserving complex references. Unlike prior DP synthesis, it targets both data-distribution fidelity and query-runtime similarity to the source. (summarized by gpt-5.6-luna on Jul 24 2026)

Paper ID: 14219
Venue: VLDB
Year: 2025
Pagerank: 5.093636e-05
Overall Rank: 10,967 | 24.76%
DOI: 10.14778/3705829.3705855

Incoming Non-self Citations Over Time

No non-self incoming citations found for this paper in this database.

Authors

1. Yunqing Ge (Shenzhen University)
2. Jianbin Qin (Shenzhen University)
3. Shuyuan Zheng (Osaka University)
4. Yongrui Zhong (Shenzhen University)
5. Bo Tang (Southern University of Science and Technology)
6. Yu-Xuan Qiu (Beijing Institute of Technology)
7. Rui Mao (Shenzhen University)
8. Ye Yuan (Beijing Institute of Technology)
9. Makoto Onizuka (Osaka University)
10. Chuan Xiao (Nagoya University; Osaka University)

BibTeX Citation

@article{ge_vldb25,
        title = {{Privacy-Enhanced Database Synthesis for Benchmark Publishing}},
        author = {Ge, Yunqing and Qin, Jianbin and Zheng, Shuyuan and Zhong, Yongrui and Tang, Bo and Qiu, Yu-Xuan and Mao, Rui and Yuan, Ye and Onizuka, Makoto and Xiao, Chuan},
        journal = {PVLDB},
        series = {{VLDB} '25},
        volume = {18},
        number = {2},
        pages = {413--425},
        doi = {10.14778/3705829.3705855},
        url = {https://doi.org/10.14778/3705829.3705855},
        year = {2025}
}

Incoming Citations (Sorted by Pagerank)

Showing 0 of 0 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank

Outgoing Citations (Sorted by Pagerank)

Showing 19 of 19 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
18	How Good Are Query Optimizers, Really?	2016	VLDB	0.00059284255
84	Learned Cardinalities: Estimating Correlated Joins with Deep Learning	2019	CIDR	0.00035838391
323	DeepDB: Learn from Data, not from Queries!	2020	VLDB	0.00021264788
388	Preventing Bad Plans by Bounding the Impact of Cardinality Estimation Errors	2009	VLDB	0.00019410042
513	NeuroCard: One Cardinality Estimator for All Tables	2021	VLDB	0.00017190574
1,122	Cardinality Estimation in DBMS: A Comprehensive Benchmark Evaluation	2022	VLDB	0.0001209124
1,144	PrivateSQL: A Differentially Private SQL Query Engine	2019	VLDB	0.00011999046
1,488	Towards Model-based Pricing for Machine Learning in a Data Marketplace	2019	SIGMOD	0.00010612416
2,436	Computing Local Sensitivities of Counting Queries with Joins	2020	SIGMOD	8.5806427e-05
2,476	Data Synthesis via Differentially Private Markov Random Fields	2021	VLDB	8.5258582e-05
2,841	AIM: An Adaptive and Iterative Mechanism for Differentially Private Synthetic Data	2022	VLDB	8.0637668e-05
3,412	Kamino: Constraint-Aware Differentially Private Data Synthesis	2021	VLDB	7.4344343e-05
3,759	Dealer: An End-to-End Model Marketplace with Differential Privacy	2021	VLDB	7.14674e-05
3,939	DPSynthesizer: Differentially Private Data Synthesizer for Privacy Preserving Data Sharing	2014	VLDB	7.008159e-05
4,106	Secure Shapley Value for Cross-Silo Federated Learning	2023	VLDB	6.8980966e-05
4,394	R2T: Instance-optimal Truncation for Differentially Private Query Evaluation with Foreign Keys	2022	SIGMOD	6.7274063e-05
4,827	PrivLava: Synthesizing Relational Data with Foreign Keys under Differential Privacy	2023	SIGMOD	6.4895329e-05
5,105	SAM: Database Generation from Query Workloads with Supervised Autoregressive Models	2022	SIGMOD	6.3628539e-05
7,358	PreFair: Privately Generating Justifiably Fair Synthetic Data	2023	VLDB	5.6344213e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	7,850	Differentially Private Data Release over Multiple Tables	2023	PODS
2	7,147	Architecting a Differentially Private SQL Engine	2019	CIDR
3	10,282	Prism: Private Relational Data Synthesis with Language Models	2026	SIGMOD
4	9,777	PBench: Workload Synthesizer with Real Statistics for Cloud Analytics Benchmarking	2025	VLDB
5	4,827	PrivLava: Synthesizing Relational Data with Foreign Keys under Differential Privacy	2023	SIGMOD
6	6,376	DProvDB: Differentially Private Query Processing with Multi-Analyst Provenance	2023	SIGMOD
7	2,300	Principled Evaluation of Differentially Private Algorithms using DPBench	2016	SIGMOD
8	7,924	PSynDB: Accurate and Accessible Private Data Generation	2019	VLDB
9	10,775	PrivPetal: Relational Data Synthesis via Permutation Relations	2025	SIGMOD
10	10,345	Benchmarking Differentially Private Tabular Data Synthesis: [Experiments & Analysis]	2026	SIGMOD