TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulation

作者：Junjie Wen, Yichen Zhu, Jinming Li, Minjie Zhu, Kun Wu, Zhiyuan Xu, Ning Liu, Ran Cheng, Chaomin Shen, Yaxin Peng, Feifei Feng, Jian Tang · 单位：Junjie Wen, Minjie Zhu, and Chaomin Shen are with East China Normal University, Shanghai 200042, China, Kun Wu is with Syracuse University, New York 13244, USA, Jinming Li and Yaxin Peng are with Shanghai University, Shanghai 201900, China · 会议/期刊：IEEE RA-L 2025 · 方向：高效动作解码 / 轻量组件 / 其他预训练策略 · 发布日期：2024-09-19