Embodied AI TopConf · ICLR2025

TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policies

ICLR2025 / Vision-Language-Action Models

视觉语言动作三维视觉感知
新标签打开 AlphaXiv新标签打开 Project
TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policies figure
AlphaXiv 中文论文页面（可滚动查看）