We propose SERUM: an intriguingly simple yet highly effective method for mark- ing images generated by diffusion models (DMs). We only add a unique watermark noise to the initial diffusion generation noise and train a lightweight detector to identify the signature of this watermark directly in the images, simplifying and unifying the strengths of prior approaches. SERUM provides robustness against any image augmentations or watermark removal attacks and is extremely efficient, all while maintaining negligible impact on image quality. In contrast to prior approaches, which are often only resilient to limited perturbations and incur sig- nificant training, injection, and detection costs, our SERUM achieves remarkable performance, with the highest true positive rate (TPR) at a 1% false positive rate (FPR) in most scenarios, along with fast injection and detection and low detector training overhead. Its decoupled architecture also seamlessly supports multiple users by embedding individualized watermarks with little interference between the marks. Overall, our method provides a practical solution to mark outputs from DMs and to reliably distinguish generated from natural images. Our code is available at https://github.com/Hubizon/SERUM.
International Conference on Learning Representations (ICLR)
2026-04-25
2026-04-07