Stanley Wei

Improved high-dimensional estimation with Langevin dynamics and stochastic weight averaging

What Makes a Reward Model a Good Teacher? An Optimization Perspective

Provable unlearning in topic modeling and downstream tasks