Evaluating Adversarial Protections for Diffusion Personalization: A Comprehensive Study

Kai Ye, Tianyi Chen, Zhen Wang

Keywords: Diffusion Models, Adversarial Perturbations, Personalized Image Generation, Privacy Protection, Robustness Evaluation

TL;DR: We develop a unified evaluation framework and provide empirical insights for selecting adversarial perturbation methods to protect diffusion model personalization.

Abstract: With the increasing adoption of diffusion models for image generation and personalization, concerns regarding privacy breaches and content misuse have become more pressing. In this study, we conduct a comprehensive comparison of eight perturbation-based protection methods—AdvDM, ASPL, FSGM, MetaCloak, Mist, PhotoGuard, SDS, and SimAC—across both portrait and artwork domains. These methods are evaluated under varying perturbation budgets, using a range of metrics to assess visual imperceptibility and protective efficacy. Our results offer practical guidance for method selection. Code is available at: https://github.com/vkeilo/DiffAdvPerturbationBench.

Submission Number: 107