How to Backdoor Diffusion Models?

by   Sheng Yen Chou, et al.

Diffusion models are state-of-the-art deep learning empowered generative models that are trained based on the principle of learning forward and reverse diffusion processes via progressive noise-addition and denoising. To gain a better understanding of the limitations and potential risks, this paper presents the first study on the robustness of diffusion models against backdoor attacks. Specifically, we propose BadDiffusion, a novel attack framework that engineers compromised diffusion processes during model training for backdoor implantation. At the inference stage, the backdoored diffusion model will behave just like an untampered generator for regular data inputs, while falsely generating some targeted outcome designed by the bad actor upon receiving the implanted trigger signal. Such a critical risk can be dreadful for downstream tasks and applications built upon the problematic model. Our extensive experiments on various backdoor attack settings show that BadDiffusion can consistently lead to compromised diffusion models with high utility and target specificity. Even worse, BadDiffusion can be made cost-effective by simply finetuning a clean pre-trained diffusion model to implant backdoors. We also explore some possible countermeasures for risk mitigation. Our results call attention to potential risks and possible misuse of diffusion models.


page 5

page 6

page 8

page 11

page 13

page 15

page 16

page 17


Diff-Instruct: A Universal Approach for Transferring Knowledge From Pre-trained Diffusion Models

Due to the ease of training, ability to scale, and high sample quality, ...

On Analyzing Generative and Denoising Capabilities of Diffusion-based Deep Generative Models

Diffusion-based Deep Generative Models (DDGMs) offer state-of-the-art pe...

A data augmentation perspective on diffusion models and retrieval

Diffusion models excel at generating photorealistic images from text-que...

VillanDiffusion: A Unified Backdoor Attack Framework for Diffusion Models

Diffusion Models (DMs) are state-of-the-art generative models that learn...

Boundary Guided Mixing Trajectory for Semantic Control with Diffusion Models

Applying powerful generative denoising diffusion models (DDMs) for downs...

GSURE-Based Diffusion Model Training with Corrupted Data

Diffusion models have demonstrated impressive results in both data gener...

Please sign up or login with your details

Forgot password? Click here to reset