Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Early Failure Detection on LIBERO Unseen
Loading...
73.8
ROC-AUC (q=0.25)
ActProbe
40.728
49.314
57.9
66.486
Jun 7, 2026
ROC-AUC (q=0.25)
Updated 1mo ago
Evaluation Results
Method
Method
Links
ROC-AUC (q=0.25)
ActProbe
Policy=OpenVLA
2026.06
73.8
ActProbe
Policy=π0
2026.06
71.4
SAFE-MLP
Policy=OpenVLA
2026.06
67.9
STAC-Single
Policy=π0
2026.06
65.2
SAFE-LSTM-TDQC
Policy=π0
2026.06
62.8
SAFE-MLP-TDQC
Policy=π0
2026.06
60.7
SAFE-LSTM
Policy=π0
2026.06
59.2
SAFE-MLP
Policy=π0
2026.06
58.9
LogpZO
Policy=π0
2026.06
58.3
SAFE-LSTM-TDQC
Policy=OpenVLA
2026.06
57.6
Cosine k-NN
Policy=π0
2026.06
57.2
SAFE-LSTM
Policy=OpenVLA
2026.06
55.9
SAFE-MLP-TDQC
Policy=OpenVLA
2026.06
52.8
LogpZO
Policy=OpenVLA
2026.06
52.6
Cosine k-NN
Policy=OpenVLA
2026.06
42
Feedback
Search any
task
Search any
task