UCSD Hao AI Lab Releases ORA v0.1, Improving Cold-Start Performance for RUA Post-Training
AI summary
UCSD Hao AI Lab has released ORA v0.1, an open runtime framework for zero-shot robot use, providing cold-start gains for RUA post-training. On RoboLab-120, the base Qwen3.8-27B scores 5.4% under the PEAR setting and 52.5% under ORA; PEAR’s RL checkpoint reaches 61.9% under ORA.
Why read this
SelectedAI score 63
The same model’s performance differences under the PEAR and ORA settings provide a direct comparison for assessing how robot agent runtime frameworks affect post-training.
Topics
Body
ORA gives RUA post-training a cold start and a free win. On RoboLab-120, base Qwen3.8-27B scored 5.4% in PEAR’s setup and 52.5% in ORA. PEAR’s RL checkpoint reached 61.9% in ORA. https://t.co/4reWMSYDsm
AI Trends