Preprint
Aug 2026
$R^3$: Training Robots to Reason in Natural Language via Reinforcement Learning
This paper introduces a simple post-training recipe that turns off-the-shelf VLMs into robotic reasoners, and suggests that free-form language reasoning can function as a test-time compute mechanism for steering low-level policies.
Lehong Wu, Yuxiao Qu, Zheyuan Hu et al.
· 0 citations