ALIEN: Analytic Latent Watermarking for Controllable Generation

TL;DR

ALIEN introduces an analytical framework for watermarking in latent diffusion models, enabling controllable, robust, and high-quality watermark embedding without intensive heuristic optimization, significantly outperforming existing methods.

Contribution

It provides the first analytical derivation of the modulation coefficient for watermarking in diffusion models, improving efficiency and robustness over prior heuristic-based approaches.

Findings

ALIEN-Q outperforms state-of-the-art by 33.1% on quality metrics.

ALIEN-R shows 14.0% improved robustness against threats.

The framework enables controllable and efficient watermark embedding.

Abstract

Watermarking is a technical alternative to safeguarding intellectual property and reducing misuse. Existing methods focus on optimizing watermarked latent variables to balance watermark robustness and fidelity, as Latent diffusion models (LDMs) are considered a powerful tool for generative tasks. However, reliance on computationally intensive heuristic optimization for iterative signal refinement results in high training overhead and local optima entrapment.To address these issues, we propose an \underline{A}na\underline{l}ytical Watermark\underline{i}ng Framework for Controllabl\underline{e} Generatio\underline{n} (ALIEN). We develop the first analytical derivation of the time-dependent modulation coefficient that guides the diffusion of watermark residuals to achieve controllable watermark embedding pattern.Experimental results show that ALIEN-Q outperforms the state-of-the-art by 33.1\% across 5 quality metrics, and ALIEN-R demonstrates 14.0\% improved robustness against generative variant and stability threats compared to the state-of-the-art across 15 distinct conditions. Code can be available at https://anonymous.4open.science/r/ALIEN/.