Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Failure Prediction on Real-robot trajectories 80 rollouts (test)
Loading...
91.2
AUC
SMD (ReconVLA)
65.096
71.873
78.65
85.427
Apr 17, 2026
AUC
Updated 3mo ago
Evaluation Results
Method
Method
Links
AUC
SMD (ReconVLA)
type=state anomaly score
2026.04
91.2
CQR (ReconVLA)
type=action uncertaint...
2026.04
75.9
A-VI
2026.04
74.6
A-AI
2026.04
74.2
A-PI
2026.04
71.9
TB-D
2026.04
71.3
TB-E
2026.04
70.5
TB-PCS
2026.04
66.2
TB-TP
2026.04
66.1
Feedback
Search any
task
Search any
task