A Locally Tokenized Generative Model for Robust Time-Series Watermarking

2026-08-20

AI Summary

该研究针对多变量时间序列水印易受攻击的问题，提出了一种局部令牌化生成模型L-VQVAE及水印方法LVQMark，以提高检测稳定性。

Abstract

Watermarking is a central tool for provenance in generative models, yet its application to multivariate time series remains hindered by reliability failures under post-editing attacks. We show that existing detectors, which rely on globally coupled re-encoding, suffer from bidirectional drift of the null distribution: post-editing attacks can shift the z-score of non-watermarked samples in either direction, invalidating clean-calibrated thresholds. We argue that this instability is a property of the re-encoding, and that reliable detection requires each recovered unit to depend only on a bounded temporal neighborhood. Guided by this principle, we propose L-VQVAE, a generative model in which each discrete token is produced from a short contiguous window, and LVQMark, a watermarking method over this token space that combines logit-bias injection with robust re-encoding for attack-time detection. Experiments on four benchmarks spanning finance, energy, and neuroimaging show that our approach preserves generation quality while stabilizing both detection power and false-positive behavior under post-editing attacks.

Problem

• Watermarking
• Time Series
• Post-editing Attacks
• Reliability

Innovation

• L-VQVAE
• LVQMark
• logit-bias injection
• robust re-encoding