James Diffenderfer

UProp: Investigating the Uncertainty Propagation of LLMs in Multi-Step Decision-Making

Trajectory Balance with Asynchrony: Decoupling Exploration and Learning for Fast, Scalable LLM Post-Training

When SNN meets ANN: Error-Free ANN-to-SNN Conversion for Extreme Edge Efficiency

LLM Unlearning Reveals a Stronger-Than-Expected Coreset Effect in Current Benchmarks