Rui Lu3 papers on this site (2023–2025)Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM ReasoningShenzhi Wang, Le Yu, Chang Gao, et al.2025Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?Yang Yue, Zhiqi Chen, Rui Lu, et al.2025AgentTuning: Enabling Generalized Agent Abilities for LLMsAohan Zeng, Mingdao Liu, Rui Lu, et al.2023