Awesome Robotics Manipulation · full_paper

Universal Pose Pretraining for Generalizable Vision-Language-Action Policies

arXiv · 2026-02-23 · Low-Level Learning-Based Action Modelling / Input Modelling / 3D Vision Language Action Models

3D 表征VLA泛化基础模型感知机器人学习
新标签打开 AlphaXiv新标签打开 Paper新标签打开 Project
Universal Pose Pretraining for Generalizable Vision-Language-Action Policies figure
AlphaXiv 中文论文页面（可滚动查看）