Elijah: Eliminating Backdoors Injected in Diffusion Models via Distribution Shift
This addresses a security vulnerability in diffusion models, which are widely used for image generation, by providing an effective defense against backdoor attacks.
The paper tackles the problem of backdoor attacks in diffusion models by proposing Elijah, a detection and removal framework that achieves close to 100% detection accuracy and reduces backdoor effects to near zero without significantly harming model utility.
Diffusion models (DM) have become state-of-the-art generative models because of their capability to generate high-quality images from noises without adversarial training. However, they are vulnerable to backdoor attacks as reported by recent studies. When a data input (e.g., some Gaussian noise) is stamped with a trigger (e.g., a white patch), the backdoored model always generates the target image (e.g., an improper photo). However, effective defense strategies to mitigate backdoors from DMs are underexplored. To bridge this gap, we propose the first backdoor detection and removal framework for DMs. We evaluate our framework Elijah on hundreds of DMs of 3 types including DDPM, NCSN and LDM, with 13 samplers against 3 existing backdoor attacks. Extensive experiments show that our approach can have close to 100% detection accuracy and reduce the backdoor effects to close to zero without significantly sacrificing the model utility.