no code implementations • 5 Feb 2024 • Yang Sui, Huy Phan, Jinqi Xiao, Tianfang Zhang, Zijie Tang, Cong Shi, Yan Wang, Yingying Chen, Bo Yuan
In this paper, for the first time, we systematically explore the detectability of the poisoned noise input for the backdoored diffusion models, an important performance metric yet little explored in the existing works.