Stanford's Real-Time EXPO-FT Brings RL to Real-Time VLA Robot Policies, Beats RTC
StanfordAILab · x · 2026-09-24
Stanford researchers (Perry Dong, Kuo-Han Hung, Dorsa Sadigh, Chelsea Finn) introduce Real-Time EXPO-FT, an RL fine-tuning framework for real-time vision-language-action policies. It tackles latency-induced distribution shift in dynamic tasks via a three-stage design: a large VLA base slowly proposes action chunks, a small edit policy revises them using observations at execution time, and a Q-function selects the best candidate on the fly. It unlocks π0.5 on tasks like ball balancing and striking, significantly outperforming RTC baselines. Paper and code are public.
Related event: Stanford's Real-Time EXPO-FT Makes VLA Robots Real-Time(2 posts)→
More from Embodied
- Homebrew robot arm build log Day 7: geartrain CAD, wiring and gripper sizing — _Stocko_ · 2026-11-11
- Perceptron AI launches EgoHome, an open benchmark for sub-second egocentric hand understanding — rohanpaul_ai · 2026-11-11
- Runway open-sources Praxis-1, turning video pretraining into real robot control — runwayml · 2026-11-11
- Unitree G1 humanoid wearing VR goggles pilots an A1 drone in land-air robot demo — CyberRobooo · 2026-11-11
- CoRL Spotlight Meta-Ctrl predicts robot plan failures, Llama 3 8B beats GPT-5.5 on VirtualHome — guyvdb · 2026-11-11
- Luna, a jewelry-style AI wearable that speaks before you ask, launches on Kickstarter — nikola_mr64990 · 2026-11-11