RL-VLA$^3$: A Flexible and Asynchronous Reinforcement Learning Framework for VLA Training
RL-VLA is the first fully asynchronous RL training framework tailored specifically for the system-level challenges of VLA training, and achieves throughput improvements of up to 85.2\% over synchronous baselines while maintaining identical sample efficiency, with scalability validated from 8 to 256 GPUs.
Haoran Sun, Yong-Jian Guo, Zhong Guan et al.
· 7 citations