Yunjuan Wang

When Does Curriculum Learning Help? A Theoretical Perspective

Backdoor Attacks in Token Selection of Attention Mechanism

Stability and Generalization of Adversarial Training for Shallow Neural Networks with Smooth Activation

On the Stability and Generalization of Meta-Learning

Robustness Guarantees for Adversarial Training on Non-Separable Data

DART: A Principled Approach to Adversarially Robust Unsupervised Domain Adaptation