mlsanityData to 7 Oct 2026

Differentially Private Synthetic Data via APIs 4: Tabular Data

arXiv 2606.08259 ↗First posted 6 Jun 2026cs.LGcited by 2

On the map: Synthetic data and self-improvement for LLMs → Privacy-preserving tabular data synthesis

Abstract (arXiv)

This paper investigates the problem of generating synthetic tabular data with differential privacy (DP) guarantees, enabling data sharing in sensitive domains. Despite extensive study, state-of-the-art methods often focus on minimizing low-order marginal query errors and overlook the challenges posed by high-order correlations. To address this gap, we extend the Private Evolution (PE) framework, originally developed for DP-compliant image and text synthesis, to tabular data. We introduce Tab-PE -- an algorithm for synthetic tabular data generation under DP constraints. Tab-PE iteratively improves a candidate dataset via an evolutionary process that leverages tabular-specialized operators to produce variations, privately scores them, and selects the highest-quality samples to retain and propagate. In contrast to the original PE, which relies on large foundation models, Tab-PE employs heuristic operators with significantly lower computational costs, making PE more practical and scalable for tabular data. Through extensive experiments on real-world and simulation datasets, we demonstrate that Tab-PE substantially outperforms prior baselines on datasets exhibiting high-order correlations. Compared to the best baseline -- AIM, Tab-PE improves classification accuracy by up to 10% while running 28 times faster.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in Privacy-preserving tabular data synthesis, one of 13 directions in Synthetic data and self-improvement for LLMs. Builds on 43 of its 55 references that are in the map. First posted on arXiv 6 Jun 2026.

Builds on

43 of 55 references.

Closest ancestors · 7

Same direction, newest first

  1. 2025POPri: Private Federated Learning using Preference-Optimized Synthetic Data23 Apr
  2. 2025Differentially Private Synthetic Data via APIs 3: Using Simulators Instead of Foundation Model8 Feb
  3. 2025Contrastive Private Data Synthesis via Weighted Multi-PLM Fusion1 Feb
  4. 2024Differentially Private Tabular Data Synthesis using Large Language Models3 Jun
  5. 2024ClavaDDPM: Multi-relational Data Synthesis with Cluster-guided Diffusion Models28 May
  6. 2024Differentially Private Synthetic Data via Foundation Model APIs 2: Text4 Mar
  7. 2024Systematic Assessment of Tabular Data Synthesis9 Feb

Other references · 36

  1. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,120
  2. 2021Deep Neural Networks and Tabular Data: A Surveycited by 489
  3. 2021Retiring Adult: New Datasets for Fair Machine Learningcited by 443
  4. 2018Assessing Generative Models via Precision and Recallcited by 401
  5. 2025TabICL: A Tabular Foundation Model for In-Context Learning on Large Datacited by 267
  6. 2018Improving the Gaussian Mechanism for Differential Privacy: Analytical Calibration and Optimal Denoisingcited by 216
  7. 2022Differentially Private Diffusion Modelscited by 118
  8. 2022Synthetic Text Generation with Differential Privacy: A Simple and Practical Recipecited by 108
  9. 2021Winning the NIST Contest: A scalable and general approach to differentially private synthetic datacited by 101
  10. 2022AIM: An Adaptive and Iterative Mechanism for Differentially Private Synthetic Datacited by 88
  11. 2021Benchmarking Differentially Private Synthetic Data Generation Algorithmscited by 80
  12. 2019Graphical-model based estimation and inference for differential privacycited by 73
24 more
  1. 2023Harnessing large-language models to generate private synthetic textcited by 65
  2. 2023Differentially Private Synthetic Data via Foundation Model APIs 1: Imagescited by 63
  3. 2021Iterative Methods for Private Synthetic Data: Unifying Framework and New Methodscited by 58
  4. 2019Diffprivlib: The IBM Differential Privacy Librarycited by 51
  5. 2023SoK: Privacy-Preserving Data Synthesiscited by 37
  6. 2022Private Synthetic Data for Multitask Learning and Marginal Queriescited by 35
  7. 2024PrE-Text: Training Language Models on Private Federated Data in the Age of LLMscited by 29
  8. 2020Kamino: Constraint-Aware Differentially Private Data Synthesiscited by 23
  9. 2016PrivTree: A Differentially Private Algorithm for Hierarchical Decompositionscited by 22
  10. 2023Generating Private Synthetic Data with Genetic Algorithmscited by 18
  11. 2023DP-TBART: A Transformer-based Autoregressive Model for Differentially Private Tabular Data Generationcited by 13
  12. 2024Privacy-preserving data release leveraging optimal transport and particle gradient descentcited by 8
  13. 2024Joint Selection: Adaptively Incorporating Public Information for Private Synthetic Datacited by 8
  14. 2025Synthetic Tabular Data: Methods, Attacks and Defensescited by 8
  15. 2025Synthesize Privacy-Preserving High-Resolution Images via Private Textual Intermediariescited by 8
  16. 2021DPSyn: Experiences in the NIST Differential Privacy Data Synthesis Challengescited by 7
  17. 2025Struct-Bench: A Benchmark for Differentially Private Structured Text Generationcited by 7
  18. 2023FLAIM: AIM-based Synthetic Data Generation in the Federated Settingcited by 6
  19. 2024Tabular Data Synthesis with Differential Privacy: A Surveycited by 6
  20. 2025Is API Access to LLMs Useful for Generating Private Synthetic Tabular Data?cited by 5
  21. 2025PCEvolve: Private Contrastive Evolution for Synthetic Dataset Generation via Few-Shot Private Data and Generative APIscited by 4
  22. 2024Differentially Private Synthetic Data Generation for Relational Databasescited by 2
  23. 2025GEM+: Scalable State-of-the-Art Private Synthetic Data with Generator Networkscited by 1
  24. 2026Privately Fine-Tuned LLMs Preserve Temporal Dynamics in Tabular Datacited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026QuanText: Protecting Dataset-Level Secrets in Textual Data Sharingcited by 0
  2. 2026PEG-Tab: Sampling-Time Record Repair and Release Control for Tabular Synthesiscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Membership Inference over Diffusion-models-based Synthetic Tabular Data
  2. 2025A Comprehensive Survey of Synthetic Tabular Data Generation
  3. 2026Creation begins with understanding: LLMs as strategy designers for privacy-preserving tabular data synthesis
  4. 2025Privacy Preserving Diffusion Models for Mixed-Type Tabular Data Generation
  5. 2025Flow Matching for Tabular Data Synthesis
  6. 2024Federated Diffusion Modeling with Differential Privacy for Tabular Data Synthesis
  7. 2024DP-TLDM: Differentially Private Tabular Latent Diffusion Model
  8. 2025Risk In Context: Benchmarking Privacy Leakage of Foundation Models in Synthetic Tabular Data Generation