Score: 0

Membership Inference over Diffusion-models-based Synthetic Tabular Data

Published: October 16, 2025 | arXiv ID: 2510.16037v1

By: Peini Cheng, Amir Bahmani

Potential Business Impact:

Protects private data when making fake data.

Business Areas:
Predictive Analytics Artificial Intelligence, Data and Analytics, Software

This study investigates the privacy risks associated with diffusion-based synthetic tabular data generation methods, focusing on their susceptibility to Membership Inference Attacks (MIAs). We examine two recent models, TabDDPM and TabSyn, by developing query-based MIAs based on the step-wise error comparison method. Our findings reveal that TabDDPM is more vulnerable to these attacks. TabSyn exhibits resilience against our attack models. Our work underscores the importance of evaluating the privacy implications of diffusion models and encourages further research into robust privacy-preserving mechanisms for synthetic data generation.

Page Count
8 pages

Category
Computer Science:
Cryptography and Security