TableMark: A Multi-bit Watermark for Synthetic Tabular Data

2026-03-13

AI Summary

This work proposes TableMark, the first multi-bit watermarking scheme tailored for synthetic tabular data, addressing the challenge of preserving watermark integrity against perturbations while simultaneously ensuring high data utility, robustness, and support for multi-user tracing. By formulating a constrained optimization framework that explicitly balances utility and robustness, TableMark integrates watermark embedding with an efficient tracing mechanism and further enhances resilience through attack-resistant strategies. Extensive experiments on four real-world datasets demonstrate that TableMark significantly outperforms existing methods, achieving strong robustness against common post-processing operations, maintaining high data utility, and enabling precise attribution across a large number of users.

Technology Category

• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust
• Machine Learning: Privacy
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Watermarking has emerged as an effective solution for copyright protection of synthetic data. However, applying watermarking techniques to synthetic tabular data presents challenges, as tabular data can easily lose their watermarks through shuffling or deletion operations. The major challenge is to provide traceability for tracking multiple users of the watermarked tabular data while maintaining high data utility and robustness (resistance to attacks). To address this, we design a multi-bit watermarking scheme TableMark that encodes watermarks into synthetic tabular data, ensuring superior traceability and robustness while maintaining high utility. We formulate the watermark encoding process as a constrained optimization problem, allowing the data owner to effectively trade off robustness and utility. Additionally, we propose effective optimization mechanisms to solve this problem to enhance the data utility. Experimental results on four widely used real-world datasets show that TableMark effectively traces a large number of users, is resilient to attacks, and preserves high utility. Moreover, TableMark significantly outperforms state-of-the-art tabular watermarking schemes.

Problem

• watermarking
• synthetic tabular data
• traceability
• robustness
• data utility

Innovation

• multi-bit watermarking
• synthetic tabular data
• constrained optimization
• robustness-utility trade-off
• traceability