SOURCE-LINKED INTELLIGENCE

DRIFT: Removing Diffusion Watermarks by Deflecting the Generative Trajectory

arXiv · AI, language, vision and robotics · article · Sep 8, 2026 · UTC

Diffusion watermarking embeds verifiable signals into the generative process and commonly verifies them by recovering trajectory-dependent evidence, making the marks robust to conventional pixel-space distortions. Existing removal attacks either regenerate along deterministic trajectories, which often preserve the watermark-bearing latent structure, or optimize every image separately. We identify the reliance on a recoverable generative trajectory as a common attack surface among the schemes we study. Based on this observation, we propose DRIFT, a black-box attack that combines partial forward

Read original source ↗ Open in workspace

recordType

paper

region

Global