Tsui-Wei Weng

10

Papers

16

Total Citations

Papers (10)

Evaluating Neuron Explanations: A Unified Framework with Sanity Checks

Interpretable Generative Models through Post-hoc Concept Bottlenecks

Interpretability-Guided Test-Time Adversarial Defense

Understanding Fixed Predictions via Confined Regions

Concept Bottleneck Large Language Models

Towards Verifying Robustness of Neural Networks Against A Family of Semantic Perturbations

Corrupting Neuron Explanations of Deep Visual Features

Higher-Order Certification For Randomized Smoothing

NeurIPS 2020arXiv

On the Equivalence between Neural Network and Support Vector Machine

NeurIPS 2021arXiv

Robust Deep Reinforcement Learning through Adversarial Loss

NeurIPS 2021arXiv