List your company

Failure Detection on VLABench (Unseen Tasks)

71.3bACCHide-and-Seek
47.48453.66759.8566.033May 29, 2026

Evaluation results

MethodLinks
Hide-and-SeekPolicy=π0.52026.0571.370.960.8
Cosine k-NNPolicy=π0.52026.0570.467.658.4
LogpZOPolicy=π0.52026.0567.462.858.1
SAFE-MLPPolicy=π0.52026.0564.163.450
PCA KMeansPolicy=π0.52026.0561.661.852
MahalanobisPolicy=π0.52026.0561.154.854.5
STACPolicy=π0.52026.0559.159.857.6
ACEPolicy=π0.52026.0559.159.555.2
EigenScorePolicy=π0.52026.0556.653.855.4
RNDPolicy=π0.52026.0555.153.750.7
Cluster EntropyPolicy=π0.52026.0552.650.649
SAFE-LSTMPolicy=π0.52026.0548.453.348.2