Failure Detection and Reasoning on Maniskill
78.8Detection Accuracy
SFT-D
Updated 4d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.02 | 78.8 | 0.644 | 74.3 | |
2026.02 | 62.5 | 0.398 | 21.2 | |
2026.02 | 54.8 | 0.268 | 11.2 | |
2026.02 | 53.8 | 0.36 | 13.3 | |
2026.02 | 44.2 | 0.359 | 20.7 | |
2026.02 | 1 | 0.286 | 4.2 |