List your company

Early Failure Detection on RoboCasa (Unseen)

76.3ROC-AUC (q=0.25)ActProbe
60.764.7568.872.85Jun 7, 2026

Evaluation results

MethodLinks
ActProbePolicy=GR00T2026.0676.3
STAC-SinglePolicy=GR00T2026.0673.8
Cosine k-NNPolicy=GR00T2026.0670.3
SAFE-LSTM-TDQCPolicy=GR00T2026.0665.9
LogpZOPolicy=GR00T2026.0665.5
SAFE-MLP-TDQCPolicy=GR00T2026.0664.7
SAFE-MLPPolicy=GR00T2026.0661.5
SAFE-LSTMPolicy=GR00T2026.0661.3