RL-VLA3: Reinforcement Learning VLA Accelerating via Full Asynchronism

作者：Preprint. Under review. RL-VLA : A Flexible, Yongjian Guo, Zhong Guan, Shuai Di, Xiaodong Bai, Jing Long, Tianyun Zhao, Luoming Xi, Hongke Zhao, Likang Wu, Xiaotie Deng, Xu Chu, Xi Xiao, Sheng Wen, Yicheng Gong, Junwu Xiong · 单位：Peking University, Tsinghua University, Tianjin University, JDT AI Infra, Swinburne University of Technology · 会议/期刊：arXiv · 日期：2026-02-05 · 来源：Low-Level Learning-Based Action Modelling / Input Modelling / 2D Vision Language Action Models with RL