| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| LIBERO-PRO-Long unseen benchmark | OpenVLA | Language Error42 | 10 | 1mo ago | |
| RLBench | CoT-VLA | Close box95 | 7 | 1mo ago | |
| Real-World Unseen | π0.5-ADV | Average Success Rate44 | 6 | 1mo ago | |
| SO-101 Real-world | Average Success Rate90 | 3 | 1mo ago | ||
| Real Robot long-horizon xArm | BAKU | Success Rate84 | 2 | 1mo ago | |
| LIBERO-10 long-horizon | BAKU | Success Rate86 | 2 | 1mo ago |