Shallow Diffuse: Robust and Invisible Watermarking through Low-Dim Subspaces in Diffusion Models

NeurIPSSpotlight2025

Authors
Wenda Li, Huijie Zhang, Qing Qu
Venue
NeurIPS 2025
Track
Spotlight

TL;DR

The widespread use of AI-generated content from diffusion models has raised significant concerns regarding misinformation and copyright infringement…

Opening excerpt from the authors’ abstract. source

Read the paper

Topics

diffusion

← All NeurIPS 2025 Spotlight papers · Browse the whole archive