4 papers · 1 filter
Why the Maximum Second Derivative of Activations Matters for Adversarial Robustness
Yunrui Yu, Hang Su, Jun Zhu
This work investigates the critical role of activation function curvature -- quantified by the maximum second derivative -- in adversarial robustness. Using the Recursi…
RCR-AF: Enhancing Model Generalization via Rademacher Complexity Reduction Activation Function
Yunrui Yu, Kafeng Wang, Hang Su +1
Despite their widespread success, deep neural networks remain critically vulnerable to adversarial attacks, posing significant risks in safety-sensitive applications. This paper in…
Theoretical Analysis of Relative Errors in Gradient Computations for Adversarial Attacks with CE Loss
Yunrui Yu, Hang Su, Cheng-zhong Xu +2
Gradient-based adversarial attacks using the Cross-Entropy (CE) loss often suffer from overestimation due to relative errors in gradient computation induced by floating-point arith…
Vidar: Embodied Video Diffusion Model for Generalist Manipulation
Yao Feng, Hengkai Tan, Xinyi Mao +5
Scaling general-purpose manipulation to new robot embodiments remains challenging: each platform typically needs large, homogeneous demonstrations, and end-to-end pixel-to-action p…