Failure Detection on Drawer Real-world
91AUCPRFAIL
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| FAILPolicy=X-VLA2026.06 | 91 | 22 | |
| DiffPolicy=X-VLA2026.06 | 90 | 25 | |
| LLMDPolicy=X-VLA2026.06 | 90 | 19 | |
| ACEPolicy=X-VLA2026.06 | 89 | 23 | |
| ACCPolicy=X-VLA2026.06 | 83 | 33 | |
| STACPolicy=π0.52026.06 | 72 | 50 | |
| LLMDPolicy=π0.52026.06 | 69 | 45 | |
| FAILPolicy=π0.52026.06 | 68 | 47 | |
| DiffPolicy=π0.52026.06 | 63 | 49 | |
| ACCPolicy=π0.52026.06 | 62 | 53 | |
| ACEPolicy=π0.52026.06 | 31 | 68 | |
| STACPolicy=X-VLA2026.06 | 18 | 71 |