1
RL with Learnable Textual Feedback: A Bilevel Approach
提出双层优化方法，让可学习的文本反馈高效引导LLM推理，解决稀疏奖励下的样本效率难题。
arXiv:2605.24547v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards can improve LLM reasoning, but learning remains sample-…