Early Failure Detection on RoboCasa (Seen)
77.4ROC-AUC (q=0.25)ActProbe
Evaluation Results
| Method | Links | |
|---|---|---|
| ActProbePolicy=GR00T2026.06 | 77.4 | |
| SAFE-MLPPolicy=GR00T2026.06 | 74.2 | |
| SAFE-LSTM-TDQCPolicy=GR00T2026.06 | 74.2 | |
| STAC-SinglePolicy=GR00T2026.06 | 73.6 | |
| SAFE-MLP-TDQCPolicy=GR00T2026.06 | 71.3 | |
| SAFE-LSTMPolicy=GR00T2026.06 | 70.9 | |
| LogpZOPolicy=GR00T2026.06 | 68.6 | |
| Cosine k-NNPolicy=GR00T2026.06 | 65.5 |