Diffusion-Guided Adversarial Perturbation Injection for Generalizable Defense Against Facial Manipulations
College of Computer Science and Technology, National Huaqiao University(华侨大学计算机科学与技术学院);Xiamen Key Laboratory of Data Security and Blockchain Technology(厦门市数据安全与区块链重点实验室);Shenzhen Key Laboratory of Media Security, Shenzhen University(深圳大学媒体安全深圳市重点实验室);School of Cyber Science and Engineering, Sichuan University(四川大学网络空间科学学院);Zhejiang Key Laboratory of Artificial Intelligence of Things (AIoT) Network and Data Security(浙江省物联网(AIoT)网络与数据安全重点实验室)
Yue Li, Linying Xue, Kaiqing Lin, Hanyu Quan, Dongdong Lin, Hui Tian, Hongxia Wang, Bin Wang
Recent advances in GAN and diffusion models have significantly improved the realism and controllability of facial deepfake manipulation, raising serious concerns regarding privacy, security, and identity misuse. Proactive defenses attempt to counter this threat by injecting adversarial perturbations into images before manipulation takes place. However, existing approaches remain limited in effectiveness due to suboptimal perturbation injection strategies and are typically designed under white-box assumptions, targeting only simple GAN-based attribute editing. These constraints hinder their applicability in practical real-world scenarios. In this paper, we propose AEGIS, the first diffusion-guided paradigm in which the AdvErsarial facial images are Generated for Identity Shielding. We observe that the limited defense capability of existing approaches stems from the peak-clipping constraint, where perturbations are forcibly truncated due to a fixed $L_\infty$-bounded. To overcome this limitation, instead of directly modifying pixels, AEGIS injects adversarial perturbations into the latent space along the DDIM denoising trajectory, thereby decoupling the perturbation magnitude from pixel-level constraints and allowing perturbations to adaptively amplify where most effective. The extensible design of AEGIS allows the defense to be expanded from purely white-box use to also support black-box scenarios through a gradient-estimation strategy. Extensive experiments across GAN and diffusion-based deepfake generators show that AEGIS consistently delivers strong defense effectiveness while maintaining high perceptual quality. In white-box settings, it achieves robust manipulation disruption, whereas in black-box settings, it demonstrates strong cross-model transferability.
DoubtProbe: Black-Box Jailbreak Defense via Structural Verification and Semantic Auditing
East China Normal University(东华大学);The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所)
Nanyang Technological University(南洋理工大学);Alibaba Group(阿里巴巴集团);East China Normal University(华东师范大学);Kyushu University(九州大学);Institute of Computing Innovation, Zhejiang University(浙江大学计算创新研究院)