Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Early Failure Detection on LIBERO (Seen)
Loading...
0.809
ROC-AUC (q=0.25)
SAFE-MLP
0.56772
0.63036
0.693
0.75564
Jun 7, 2026
ROC-AUC (q=0.25)
Updated 1mo ago
Evaluation Results
Method
Method
Links
ROC-AUC (q=0.25)
SAFE-MLP
Policy=OpenVLA
2026.06
0.809
ActProbe
Policy=π0
2026.06
0.801
SAFE-LSTM-TDQC
Policy=π0
2026.06
0.796
SAFE-LSTM
Policy=π0
2026.06
0.782
SAFE-MLP
Policy=π0
2026.06
0.773
SAFE-MLP-TDQC
Policy=π0
2026.06
0.766
ActProbe
Policy=OpenVLA
2026.06
0.763
SAFE-LSTM
Policy=OpenVLA
2026.06
0.757
LogpZO
Policy=π0
2026.06
0.731
Cosine k-NN
Policy=OpenVLA
2026.06
0.721
Cosine k-NN
Policy=π0
2026.06
0.689
SAFE-LSTM-TDQC
Policy=OpenVLA
2026.06
0.675
SAFE-MLP-TDQC
Policy=OpenVLA
2026.06
0.657
STAC-Single
Policy=π0
2026.06
0.653
LogpZO
Policy=OpenVLA
2026.06
0.577
Feedback
Search any
task
Search any
task