Reinforcement Learning

Scalable Embodied Critic Agent — 一个具身 RL 的终极范式设想