Robust task completion on OGBench Cube
94.29Safety
SLS2
Updated 1mo ago
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
2026.06 | 94.29 | 97.14 | 94.29 | 0.359 | |
2026.06 | 30.56 | 86.11 | 27.78 | 0.345 | |
2026.06 | 25 | 91.67 | 22.22 | 0.349 | |
2026.06 | 0 | 88.57 | 0 | - | |
2026.06 | 0 | 85.71 | 0 | - | |
2026.06 | 0 | 82.86 | 0 | - |