TAB-DRW: A DFT-based Robust Watermark for Generative Tabular Data
Yizhou Zhao, Xiang Li, Peter Song, Qi Long, Weijie Su
TL;DR
This work addresses the challenge of tracing provenance in high-fidelity synthetic tabular data by introducing TAB-DRW, a post-editing watermarking method that embeds signals in the frequency domain after Yeo-Johnson transformation and standardization. It introduces a novel rank-based pseudorandom bit generation scheme enabling efficient, row-wise retrieval without storing per-row bits, and supports both hard and soft imaginary-part modifications of the DFT to balance fidelity and detectability. The paper provides theoretical distortion and robustness guarantees and validates the approach on five real datasets, demonstrating strong detectability, robustness to post-processing and adaptive attacks, and high data utility. A privacy-enhanced variant enables multi-key deployments with low false-positive risk, highlighting TAB-DRW’s practicality for provenance of generative tabular data in real-world settings.
Abstract
The rise of generative AI has enabled the production of high-fidelity synthetic tabular data across fields such as healthcare, finance, and public policy, raising growing concerns about data provenance and misuse. Watermarking offers a promising solution to address these concerns by ensuring the traceability of synthetic data, but existing methods face many limitations: they are computationally expensive due to reliance on large diffusion models, struggle with mixed discrete-continuous data, or lack robustness to post-modifications. To address them, we propose TAB-DRW, an efficient and robust post-editing watermarking scheme for generative tabular data. TAB-DRW embeds watermark signals in the frequency domain: it normalizes heterogeneous features via the Yeo-Johnson transformation and standardization, applies the discrete Fourier transform (DFT), and adjusts the imaginary parts of adaptively selected entries according to precomputed pseudorandom bits. To further enhance robustness and efficiency, we introduce a novel rank-based pseudorandom bit generation method that enables row-wise retrieval without incurring storage overhead. Experiments on five benchmark tabular datasets show that TAB-DRW achieves strong detectability and robustness against common post-processing attacks, while preserving high data fidelity and fully supporting mixed-type features.