Spherical Watermark: Encryption-Free, Lossless Watermarking for Diffusion Models

ICLROral2026

Authors
Xiaoxiao Hu, Jiaqi Jin, Sheng Li, Wanli Peng, Xinpeng Zhang, Zhenxing Qian
Affiliation
Fudan University
Venue
ICLR 2026
Track
Oral

TL;DR

Employing a novel spherical mapping mechanism, we propose a novel lossless watermarking scheme for text-to-image diffusion models.

Opening excerpt from the authors’ abstract. source

Read the paper

Topics

text-to-image diffusion

← All ICLR 2026 Oral papers · Browse the whole archive