2

Forget Less, Retain More: A Lightweight Regularizer for Rehearsal-Based Continual Learning

We leverage a lightweight regularizer to reduce forgetting in continual learning.

Test-Time Adaptation with Source Based Auxiliary Tasks

We leverage source-based auxiliary tasks to accelerate and enhance adaptation.

Generalizability of Adversarial Robustness Under Distribution Shifts

We study the generalizability of adversarial robustness to unseen domains.

ANCER: Anisotropic Certification via Sample-wise Volume Maximization

We extend randomized smoothing to anisotropic distributions and optimize the anisotropic smoothing parameters per input point in randomized smoothing.

On the Decision Boundaries of Neural Networks: A Tropical Geometry Perspective

We characterizes the decision boundaries of neural networks using tropical geometry, and leverage this characterization into several applications.