Secure Seed-Based Multi-bit Watermarking for Diffusion Models from First Principles
For researchers in generative model watermarking, this work provides a theoretical foundation and a generalizable method with guaranteed trade-offs, replacing empirical reliance.
The paper addresses the lack of rigorous theoretical evaluation for seed-based watermarking in diffusion models. It introduces a formal framework based on security, robustness, and fidelity, and proposes SSB, a method that achieves any trade-off on the characteristic surface, eliminating the need for costly empirical evaluations.
The rapid emergence of generative image models has led to the development of specialized watermarking techniques, particularly in-generation methods such as seed-based embedding. However, current evaluations in this area remain largely empirical, making them heavily reliant on the specific model architectures used for generation and inversion. This prevents any clear conclusion on the performance of any method, especially regarding security, for which a rigorous definition is lacking. Against this approach, we argue that the effectiveness of a watermarking scheme should be established purely through a thorough theoretical analysis. This is enabled by decoupling the model-dependent part from the actual decision mechanism of the watermarking system. Using this decoupling, we introduce a formal evaluation framework based on security, robustness, and fidelity. This allows precise comparisons between watermarking systems through a characteristic surface representing the trade-off between these three quantities, independent of any generative model. Based on this framework, we propose SSB, a novel watermarking method that generalizes previous seed-based methods by allowing to reach any security-robustness-fidelity regime on its characteristic surface. This work opens the door to the design of modern watermarking systems with theoretical guarantees that do not necessitate any costly empirical evaluations.