AI 进化论：如何让机器像高手一样思考？

[LG] TreeRL: LLM Reinforcement Learning with On-Policy Tree Search

[Tsinghua University & California Institute of Technology]

arxiv.org