Awesome Robotics Manipulation · full_paper

ActiveVLA: Injecting Active Perception into Vision-Language-Action Models for Precise 3D Robotic Manipulation

arXiv · 2026-01-13 · Low-Level Learning-Based Action Modelling / Input Modelling / 3D Vision Language Action Models

3D 表征VLA感知机器人学习操作
新标签打开 AlphaXiv新标签打开 Paper新标签打开 Code
ActiveVLA: Injecting Active Perception into Vision-Language-Action Models for Precise 3D Robotic Manipulation figure
AlphaXiv 中文论文页面（可滚动查看）