Failure Detection on Kitchen Real-world
100AUCPR (PR)ACC
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ACCPolicy=X-VLA2026.06 | 100 | 25 | |
| LLMDPolicy=X-VLA2026.06 | 100 | 20 | |
| FAILPolicy=X-VLA2026.06 | 100 | 21 | |
| ACCPolicy=π0.52026.06 | 99 | 17 | |
| ACEPolicy=X-VLA2026.06 | 98 | 24 | |
| DiffPolicy=X-VLA2026.06 | 98 | 21 | |
| ACEPolicy=π0.52026.06 | 95 | 20 | |
| STACPolicy=π0.52026.06 | 92 | 26 | |
| DiffPolicy=π0.52026.06 | 75 | 30 | |
| FAILPolicy=π0.52026.06 | 74 | 28 | |
| LLMDPolicy=π0.52026.06 | 62 | 33 | |
| STACPolicy=X-VLA2026.06 | 30 | 55 |