|国家预印本平台
首页|Sample-Specific Noise Injection For Diffusion-Based Adversarial Purification

Sample-Specific Noise Injection For Diffusion-Based Adversarial Purification

Sample-Specific Noise Injection For Diffusion-Based Adversarial Purification

来源:Arxiv_logoArxiv
英文摘要

Diffusion-based purification (DBP) methods aim to remove adversarial noise from the input sample by first injecting Gaussian noise through a forward diffusion process, and then recovering the clean example through a reverse generative process. In the above process, how much Gaussian noise is injected to the input sample is key to the success of DBP methods, which is controlled by a constant noise level $t^*$ for all samples in existing methods. In this paper, we discover that an optimal $t^*$ for each sample indeed could be different. Intuitively, the cleaner a sample is, the less the noise it should be injected, and vice versa. Motivated by this finding, we propose a new framework, called Sample-specific Score-aware Noise Injection (SSNI). Specifically, SSNI uses a pre-trained score network to estimate how much a data point deviates from the clean data distribution (i.e., score norms). Then, based on the magnitude of score norms, SSNI applies a reweighting function to adaptively adjust $t^*$ for each sample, achieving sample-specific noise injections. Empirically, incorporating our framework with existing DBP methods results in a notable improvement in both accuracy and robustness on CIFAR-10 and ImageNet-1K, highlighting the necessity to allocate distinct noise levels to different samples in DBP methods. Our code is available at: https://github.com/tmlr-group/SSNI.

Yuhao Sun、Jiacheng Zhang、Zesheng Ye、Chaowei Xiao、Feng Liu

计算技术、计算机技术

Yuhao Sun,Jiacheng Zhang,Zesheng Ye,Chaowei Xiao,Feng Liu.Sample-Specific Noise Injection For Diffusion-Based Adversarial Purification[EB/OL].(2025-06-06)[2025-07-23].https://arxiv.org/abs/2506.06027.点此复制

评论