| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Human3.6 | SA-GCN | MMD (Average)0.146 | 16 | 5mo ago | |
| COKE | COLM | METEOR0.342 | 13 | 5mo ago | |
| LIBERO | OpenVLA-OFT | Latency (s)0.031 | 10 | 1mo ago | |
| Do-Undo (evaluation) | DINO-R89 | 6 | 5mo ago | ||
| Hopper Hard | Flow | Safety Violation Rate0.005 | 5 | 1mo ago | |
| Walker2d Simple | PolyFlow | Safety100 | 5 | 1mo ago | |
| VIMA-Bench | VIMA | Success Rate72.6 | 5 | 5mo ago | |
| Overall | Orca | Rule-based Score32.4 | 4 | 26d ago | |
| Object OOD | π0.5 | Rule-based31.2 | 4 | 26d ago | |
| Environment OOD | Orca | Rule-based Score36.6 | 4 | 26d ago | |
| LIBERO simulation | VAG | Spatial Success Rate70 | 3 | 3mo ago | |
| Table Tennis Action Generation | SpikePingpong | Inference Time (ms)0.407 | 3 | 5mo ago | |
| Human Evaluation 48 instances (test) | Do-Undo | Preference Rate58.3 | 3 | 5mo ago | |
| ESC | COLM | Overall Score4.82 | 3 | 5mo ago |