List your company

Early Failure Detection on RoboCasa S (Seen)

74.9ROC-AUC (q=0.25)SAFE-LSTM
55.55660.57865.670.622Jun 7, 2026

Evaluation results

MethodLinks
SAFE-LSTMPolicy=π0.52026.0674.9
SAFE-MLP-TDQCPolicy=π0.52026.0674.8
SAFE-LSTM-TDQCPolicy=π0.52026.0673.9
SAFE-MLPPolicy=π0.52026.0673.7
LogpZOPolicy=π0.52026.0670.7
ActProbePolicy=π0.52026.0670
STAC-SinglePolicy=π0.52026.0662.3
Cosine k-NNPolicy=π0.52026.0656.3