Wenkai Yang

LaSeR: Reinforcement Learning with Last-Token Self-Rewarding

DeepCritic: Deliberate Critique with Large Language Models

Super(ficial)-alignment: Strong Models May Deceive Weak Models in Weak-to-Strong Generalization

Towards Thinking-Optimal Scaling of Test-Time Compute for LLM Reasoning

Learning to Focus: Causal Attention Distillation via Gradient‐Guided Token Pruning