一作ICLR 2026Poster· Reinforcement Learning / Multi-agent
Negotiated Reasoning: On Provably Addressing Relative Over-Generalization
Junjie Sheng · Yantian Wang · Bo Jin · Hongyuan Zha · Jun Wang · Wenhao Li · Xiangfeng Wang
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Junjie Sheng · Yantian Wang · Bo Jin · Hongyuan Zha · Jun Wang · Wenhao Li · Xiangfeng Wang
Xidong Yang · Wenhao Li · Junjie Sheng · Yun Hua · Haosheng Chen · Chuyun Shen · Xiangfeng Wang
Reinforcement learning (RL) remains fundamentally limited by poor data efficiency and weak generalization. Prior episodic RL methods attempt to alleviate this via external memory modules, yet they suffer from two key limitations: a representation bottleneck caused by shallow encoders, and a retri...