List your company

Failure Detection on LIBERO unseen

0.097Brier ScoreSAFE-RNN-TDQC (Ours)
0.064920.281460.4980.71454Apr 22, 2026

Evaluation results

MethodLinks
SAFE-RNN-TDQC (Ours)VLA Model=π02026.040.097
SAFE-RNN-TDQC (Ours)VLA Model=UniVLA2026.040.1
RNN-TDQC (Ours)VLA Model=UniVLA2026.040.107
SAFE-MLP-TDQC (Ours)VLA Model=π02026.040.128
SAFE-MLP-TDQC (Ours)VLA Model=UniVLA2026.040.131
SAFE-MLP BCEVLA Model=π02026.040.137
RNN-TDQC (Ours)VLA Model=π0-FAST2026.040.141
SAFE-RNNVLA Model=π0-FAST2026.040.148
SAFE-MLP-TDQC (Ours)VLA Model=π0-FAST2026.040.15
RNN-BCEVLA Model=UniVLA2026.040.152
SAFE-MLP BCEVLA Model=UniVLA2026.040.158
RNN-BCEVLA Model=π0-FAST2026.040.158
SAFE-RNNVLA Model=UniVLA2026.040.162
SAFE-MLP BCEVLA Model=π0-FAST2026.040.162
SAFE-RNN-TDQC (Ours)VLA Model=π0-FAST2026.040.163
SAFE-RNNVLA Model=π02026.040.172
RNN-TDQC (Ours)VLA Model=OpenVLA2026.040.197
RNN-BCEVLA Model=OpenVLA2026.040.206
SAFE-RNN-TDQC (Ours)VLA Model=OpenVLA2026.040.218
Avg prob.VLA Model=π0-FAST2026.040.218
Avg entropyVLA Model=π0-FAST2026.040.222
SAFE-MLP-TDQC (Ours)VLA Model=OpenVLA2026.040.229
SAFE-MLP BCEVLA Model=OpenVLA2026.040.231
Running Avg prob.VLA Model=π0-FAST2026.040.238
SAFE-RNNVLA Model=OpenVLA2026.040.255
Running Avg entropyVLA Model=π0-FAST2026.040.264
Running Avg entropyVLA Model=OpenVLA2026.040.273
Avg entropyVLA Model=OpenVLA2026.040.313
Max prob.VLA Model=π0-FAST2026.040.318
Running Avg entropyVLA Model=UniVLA2026.040.33
Running Avg prob.VLA Model=OpenVLA2026.040.356
Avg prob.VLA Model=OpenVLA2026.040.364
Max prob.VLA Model=OpenVLA2026.040.39
Avg entropyVLA Model=UniVLA2026.040.391
Avg prob.VLA Model=UniVLA2026.040.532
Running Avg prob.VLA Model=UniVLA2026.040.544
Max prob.VLA Model=UniVLA2026.040.899