Failure Detection on Mixer Real-world
98AUCPRLLMD
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| LLMDPolicy=X-VLA2026.06 | 98 | 11 | |
| ACCPolicy=X-VLA2026.06 | 97 | 14 | |
| FAILPolicy=X-VLA2026.06 | 97 | 11 | |
| DiffPolicy=X-VLA2026.06 | 95 | 14 | |
| ACEPolicy=X-VLA2026.06 | 93 | 16 | |
| ACEPolicy=π0.52026.06 | 93 | 24 | |
| ACCPolicy=π0.52026.06 | 85 | 28 | |
| FAILPolicy=π0.52026.06 | 79 | 27 | |
| LLMDPolicy=π0.52026.06 | 77 | 28 | |
| STACPolicy=π0.52026.06 | 70 | 43 | |
| DiffPolicy=π0.52026.06 | 61 | 45 | |
| STACPolicy=X-VLA2026.06 | 46 | 34 |