Research on learning through interaction, reward optimization, policy learning, and decision-making AI systems.
Zhicong Li, Lingjie Jiang, Yulan Hu +7 more
Critique-guided reinforcement learning (RL) has emerged as a powerful paradigm for training LLM agents by augmenting sparse outcome rewards with natural-language feedback. However, current methods often rely on static or offline critic models, which fail to adapt as the policy evolves. In on-policy ...
Qiang Zhang, Boli Chen, Fanrui Zhang +14 more
Reinforcement learning has substantially improved the performance of LLM agents on tasks with verifiable outcomes, but it still struggles on open-ended agent tasks with vast solution spaces (e.g., complex travel planning). Due to the absence of objective ground-truth for these tasks, current RL algo...
Qingyu Ren, Qianyu He, Jingwen Chang +6 more
Instruction-following is critical for large language models, but real-world instructions often contain logical structures such as sequential dependencies and conditional branching. Existing methods typically construct datasets with parallel constraints and optimize average rewards, ignoring logical ...
Caijun Xu, Changyi Xiao, Zhongyuan Peng +2 more
Reinforcement learning (RL) offers a principled way to enhance the reasoning capabilities of large language models, yet its effectiveness hinges on training signals that remain informative as models evolve. In practice, RL progress often slows when task difficulty becomes poorly aligned with model c...
