88 citations · 102 across the 10 of their papers we have counts for
7 papers · 1 filter
Reducing Adversarial Training Cost with Gradient Approximation
Huihui Gong
Deep learning models have achieved state-of-the-art performances in various domains, while they are vulnerable to the inputs with well-crafted but small perturbations, which are na…
Parameter-Saving Adversarial Training: Reinforcing Multi-Perturbation Robustness via Hypernetworks
Huihui Gong, Minjing Dong, Siqi Ma +3
Adversarial training serves as one of the most popular and effective methods to defend against adversarial perturbations. However, most defense mechanisms only consider a single ty…
Stealthy Physical Masked Face Recognition Attack via Adversarial Style Optimization
Huihui Gong, Minjing Dong, Siqi Ma +3
Deep neural networks (DNNs) have achieved state-of-the-art performance on face recognition (FR) tasks in the last decade. In real scenarios, the deployment of DNNs requires taking…
Alpha-Clustering in Nuclei and Its Impact on Nuclear Symmetry Energy
Shuo Yang, Ruijia Li, Chang Xu
Nuclear symmetry energy is a fundamental quantity currently under intense investigation in both nuclear physics and astrophysics. The {\it softness} or {\it stiffness} of symmetry…
GPT Self-Supervision for a Better Data Annotator
Xiaohuan Pei, Yanxi Li, Chang Xu
The task of annotating data into concise summaries poses a significant challenge across various domains, frequently requiring the allocation of significant time and specialized kno…
Two-in-one Knowledge Distillation for Efficient Facial Forgery Detection
Chuyang Zhou, Jiajun Huang, Daochang Liu +4
Facial forgery detection is a crucial but extremely challenging topic, with the fast development of forgery techniques making the synthetic artefact highly indistinguishable. Prior…
Anti-Compression Contrastive Facial Forgery Detection
Jiajun Huang, Xinqi Zhu, Chengbin Du +3
Forgery facial images and videos have increased the concern of digital security. It leads to the significant development of detecting forgery data recently. However, the data, espe…