mlsanityData to 7 Oct 2026

Differentially Private Paired Table-Image Multimodal Synthesis

arXiv 2609.00708 ↗First posted 1 Sep 2026cs.CR cs.AI cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Differentially private (DP) synthesis has been extensively studied for tabular and image data separately, yet many real-world datasets contain images paired with multivariate tabular records. Synthesizing such data is particularly challenging under DP, as the two modalities favor different private learning mechanisms while their dependence must also be preserved. To address this challenge, we propose DP-TabImage, a modality-specialized framework for private paired synthesis. DP-TabImage instantiates the factorization p(x,y)=pT(y)pI(x;|;y) using a private Probabilistic Graphical Model for the multivariate table distribution and a table-conditioned diffusion model trained with DP-SGD for the conditional image distribution. To facilitate conditional learning under clipped and noisy gradients, we further pretrain the model on private table-image prototypes, pairing privately constructed attribute-conditioned images with tabular vectors derived from the already private tabular model at no additional privacy cost. Experiments on three real-world datasets show that DP-TabImage achieves a strong balance among tabular fidelity, image fidelity, and cross-modal alignment. Our analysis further reveals that visual warm-up primarily improves marginal image fidelity, whereas aligned table-image warm-up is critical for improving cross-modal correspondence. Our source code is available in the GitHub repository, https://github.com/KaiChen9909/TabImageSyn.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 33 of its 43 references that are in the map. First posted on arXiv 1 Sep 2026.

Builds on

33 of 43 references.

References · 33

  1. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  2. 2016Deep Learning with Differential Privacycited by 2,682
  3. 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,430
  4. 2017ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseasescited by 969
  5. 2019Modeling Tabular data using Conditional GANcited by 695
  6. 2017Renyi Differential Privacycited by 643
  7. 2022TabDDPM: Modelling Tabular Data with Diffusion Modelscited by 373
  8. 2018Differentially Private Generative Adversarial Networkcited by 235
  9. 2018Subsampled R'enyi Differential Privacy and Analytical Moments Accountantcited by 234
  10. 2020DP-CGAN: Differentially Private Synthetic Data and Label Generationcited by 118
  11. 2022Differentially Private Diffusion Modelscited by 118
  12. 2020GS-WGAN: A Gradient-Sanitized Approach for Learning Differentially Private Generatorscited by 117
21 more
  1. 2022A Comprehensive Survey of Image Augmentation Techniques for Deep Learningcited by 95
  2. 2022AIM: An Adaptive and Iterative Mechanism for Differentially Private Synthetic Datacited by 88
  3. 2023Differentially Private Diffusion Models Generate Useful Synthetic Imagescited by 85
  4. 2021Benchmarking Differentially Private Synthetic Data Generation Algorithmscited by 80
  5. 2020DP-MERF: Differentially Private Mean Embeddings with Random Features for Practical Privacy-Preserving Data Generationcited by 78
  6. 2019Graphical-model based estimation and inference for differential privacycited by 73
  7. 2023Differentially Private Synthetic Data via Foundation Model APIs 1: Imagescited by 63
  8. 2021Iterative Methods for Private Synthetic Data: Unifying Framework and New Methodscited by 58
  9. 2017Variational Approaches for Auto-Encoding Generative Adversarial Networkscited by 52
  10. 2020New Oracle-Efficient Algorithms for Private Synthetic Data Releasecited by 43
  11. 2022Private Synthetic Data for Multitask Learning and Marginal Queriescited by 35
  12. 2024Differentially Private Fine-Tuning of Diffusion Modelscited by 18
  13. 2023Differentially Private Neural Tangent Kernels for Privacy-Preserving Data Generationcited by 15
  14. 2025Benchmarking Differentially Private Tabular Data Synthesiscited by 13
  15. 2023PrivImage: Differentially Private Synthetic Image Generation using Diffusion Models with Semantic-Aware Pretrainingcited by 12
  16. 2025DPImageBench: A Unified Benchmark for Differentially Private Image Synthesiscited by 11
  17. 2024On Differentially Private 3D Medical Image Synthesis with Controllable Latent Diffusion Modelscited by 7
  18. 2025From Easy to Hard: Building a Shortcut for Differentially Private Image Synthesiscited by 6
  19. 2023Method for Generating Synthetic Data Combining Chest Radiography Images with Tabular Clinical Information Using Dual Generative Modelscited by 1
  20. 2025MetaVoxel: Joint Diffusion Modeling of Imaging and Clinical Metadatacited by 1
  21. 2026Multimodal synthesis of MRI and tabular data with diffusion in a joint latent space via cross-attentioncited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024DP-TLDM: Differentially Private Tabular Latent Diffusion Model
  2. 2026Differentially Private Synthetic Data via APIs 4: Tabular Data
  3. 2025Privacy Preserving Diffusion Models for Mixed-Type Tabular Data Generation
  4. 2025Winning the MIDST Challenge: New Membership Inference Attacks on Diffusion Models for Tabular Data Synthesis
  5. 2023DP-LDMs: Differentially Private Latent Diffusion Models
  6. 2026Differentially Private Multimodal In-Context Learning
  7. 2026Protecting Sensitive Data in Image Synthesis via PAC-Private Adaptation for Diffusion Models
  8. 2024Differentially Private Tabular Data Synthesis using Large Language Models