OPTML-Group/Diffusion-MU-Attack

The official implementation of ECCV'24 paper "To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images ... For Now". This work introduces one fast and effective attack method to evaluate the harmful-content generation ability of safety-driven unlearned diffusion models.

Language: Python - Size: 11.9 MB - Last synced at: 4 months ago - Pushed at: 4 months ago - Stars: 71 - Forks: 3