Transferable Multi-Bit Watermarking Across Frozen Diffusion Models via Latent Consistency Bridges
This addresses the need for efficient and flexible watermarking in diffusion models, offering a plug-and-play solution that is incremental over existing methods by combining advantages of sampling-based and fine-tuning-based approaches.
The paper tackled the problem of watermarking diffusion models for provenance and accountability, proposing DiffMark which enables single-pass multi-bit detection with a 45x speedup (16.4 ms vs. sampling-based methods) and cross-model transferability without per-model fine-tuning.
As diffusion models (DMs) enable photorealistic image generation at unprecedented scale, watermarking techniques have become essential for provenance establishment and accountability. Existing methods face challenges: sampling-based approaches operate on frozen models but require costly $N$-step Denoising Diffusion Implicit Models (DDIM) inversion (typically N=50) for zero-bit-only detection; fine-tuning-based methods achieve fast multi-bit extraction but couple the watermark to a specific model checkpoint, requiring retraining for each architecture. We propose DiffMark, a plug-and-play watermarking method that offers three key advantages over existing approaches: single-pass multi-bit detection, per-image key flexibility, and cross-model transferability. Rather than encoding the watermark into the initial noise vector, DiffMark injects a persistent learned perturbation $δ$ at every denoising step of a completely frozen DM. The watermark signal accumulates in the final denoised latent $z_0$ and is recovered in a single forward pass. The central challenge of backpropagating gradients through a frozen UNet without traversing the full denoising chain is addressed by employing Latent Consistency Models (LCM) as a differentiable training bridge. This reduces the number of gradient steps from 50 DDIM to 4 LCM and enables a single-pass detection at 16.4 ms, a 45x speedup over sampling-based methods. Moreover, by this design, the encoder learns to map any runtime secret to a unique perturbation at inference time, providing genuine per-image key flexibility and transferability to unseen diffusion-based architectures without per-model fine-tuning. Although achieving these advantages, DiffMark also maintains competitive watermark robustness against distortion, regeneration, and adversarial attacks.