// COMPANY
Stanford University
// OVERVIEW
Stanford University research group behind the OpenVLA open-source vision-language-action model.
Discussed in
- Self-Adaptive VLA for Robust Robot Deployment
- The Embodiment Gap in Robot Foundation Models
- 📵 No baggage
- ⏸️ OpenAI hits pause
- Demystifying When and Why VLAs Fail in Contact-Rich Tasks and How to Fix Them
- Masked Visual Actions for Unified World Modeling
- World Models, Explained
- 146. 对Physical Intelligence柯丽一鸣4小时访谈：Pi的开源模型研究，机器人的江湖、族谱与主角
- Improving Robotic Generalist Policies via Flow Reversal Steering
// RECENT MENTIONS
// SIGNALS
50 SIGNALS
01
product·arXiv Physical AI·OCTOBER 9, 2026
“The paper introduces a policy formulation that conditions on observation capture times and action execution times, rather than relying on sequence position as a proxy for time.”
Source→02
mention·arXiv Physical AI·OCTOBER 9, 2026
“Senior author and head of REALab, a prominent researcher in physical AI, robotics, and manipulation.”
Source→03
mention·arXiv Physical AI·OCTOBER 9, 2026
AI-extracted from podcast / newsletter / paper summaries. May contain errors.