Submitted by Hejian Sang 6 TRIAGE: Role-Typed Credit Assignment for Agentic Reinforcement Learning LinkedIn 1
Submitted by Hejian Sang 6 Overconfident Errors Need Stronger Correction: Asymmetric Confidence Penalties for Reinforcement Learning LinkedIn 2
Submitted by Yaochen Zhu 18 SemCoT: Accelerating Chain-of-Thought Reasoning through Semantically-Aligned Implicit Tokens LinkedIn 21 2