Awesome Robotics Manipulation · full_paper

OmniStream: Mastering Perception, Reconstruction and Action in Continuous Streams

作者：Yibin Yan, Jilan Xu, Shangzhe Di, Haoning Wu, Weidi Xie · 单位：School of Artificial Intelligence, SJTU, Shanghai Innovation Institute, VGG, Oxford · 会议/期刊：arXiv · 日期：2026-03-12 · 来源：Low-Level Learning-Based Action Modelling / Input Modelling / 3D Vision Language Action Models

三维表征视觉语言动作感知机器人学习
新标签打开 AlphaXiv新标签打开 Paper新标签打开 Code
OmniStream: Mastering Perception, Reconstruction and Action in Continuous Streams figure
AlphaXiv 中文论文页面（可滚动查看）