A cross-embodiment study of OpenVLA-7B

Summary

We rolled OpenVLA-7B out on Franka, UR5e, xArm-7, and an ALOHA-style bimanual rig, on six tasks, ten seeds each. Task-success ranges from 0.71 (Franka) to 0.34 (ALOHA). We break down where the variance comes from and what it implies for zero-shot embodiment transfer.