MUSE: Model-Agnostic Tabular Watermarking via Multi-Sample Selection

TL;DR

MUSE is a novel watermarking method for tabular generative models that uses multi-sample selection and scoring to embed watermarks without relying on invertibility, achieving high detectability and robustness.

Contribution

MUSE introduces a model-agnostic watermarking approach for tabular data that leverages multi-sample selection and theoretical analysis for calibration, outperforming existing methods.

Findings

Achieves 81-89% reduction in distortion rates on fidelity metrics.

Attains a 1.0 TPR@0.1%FPR detection rate.

Demonstrates robustness against various attacks.

Abstract

We introduce MUSE, a watermarking algorithm for tabular generative models. Previous approaches typically leverage DDIM invertibility to watermark tabular diffusion models, but tabular diffusion models exhibit significantly poorer invertibility compared to other modalities, compromising performance. Simultaneously, tabular diffusion models require substantially less computation than other modalities, enabling a multi-sample selection approach to tabular generative model watermarking. MUSE embeds watermarks by generating multiple candidate samples and selecting one based on a specialized scoring function, without relying on model invertibility. Our theoretical analysis establishes the relationship between watermark detectability, candidate count, and dataset size, allowing precise calibration of watermarking strength. Extensive experiments demonstrate that MUSE achieves state-of-the-art watermark detectability and robustness against various attacks while maintaining data quality, and remains compatible with any tabular generative model supporting repeated sampling, effectively addressing key challenges in tabular data watermarking. Specifically, it reduces the distortion rates on fidelity metrics by 81-89%, while achieving a 1.0 TPR@0.1%FPR detection rate. Implementation of MUSE can be found at https://github.com/fangliancheng/MUSE.