Reinforcement learning has become the foundational paradigm for enhancing the reasoning capabilities of LLMs. Yet the dominant approach, Proximal Policy Opti…
机构:清华AIR
来源:arXiv 2610.13015 | AI4Papers 论文推荐平台