Awesome Robotics Manipulation · full_paper

VLA2: Empowering Vision-Language-Action Models with an Agentic Framework for Unseen Concept Manipulation

arXiv · 2025-10-16 · Bottlenecks / Generalization / Agent

VLA泛化操作
新标签打开 AlphaXiv新标签打开 Paper新标签打开 Project
VLA2: Empowering Vision-Language-Action Models with an Agentic Framework for Unseen Concept Manipulation figure
AlphaXiv 中文论文页面（可滚动查看）