Awesome Robotics Manipulation · full_paper

DepthCache: Depth-Guided Training-Free Visual Token Merging for Vision-Language-Action Model Inference

arXiv · 2026-03-11 · Low-Level Learning-Based Action Modelling / Input Modelling / 3D Vision Language Action Models

3D 表征VLA感知机器人学习
新标签打开 AlphaXiv新标签打开 Paper
DepthCache: Depth-Guided Training-Free Visual Token Merging for Vision-Language-Action Model Inference figure
AlphaXiv 中文论文页面（可滚动查看）