Uncertainty Calibration
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
5.2RMS Calibration Error
35
Feb 26, 2026
0.072Brier Score
28
Jun 4, 2026
0.81Expected Calibration Error (ECE)
28
Apr 15, 2026
5.97ECE
27
Feb 26, 2026
29.98ECE
27
Feb 26, 2026
0.04ECE
27
Feb 26, 2026
0.009Kuiper
24
Feb 26, 2026
0.037Kuiper
24
Feb 26, 2026
0.696AUROC
18
Feb 26, 2026
0.672AUROC
18
Feb 26, 2026
78.7AUROC
18
Feb 26, 2026
0.861AUROC
18
Feb 26, 2026
95.1Empirical Coverage
10
May 13, 2026
0.066AUSE
8
Mar 19, 2026
13.6ECE
8
Feb 26, 2026
18.4ECE
8
Feb 26, 2026
0.146ECE
8
Feb 26, 2026
0.205ECE
8
Feb 26, 2026
0.168ECE
8
Feb 26, 2026
0.17Weighted ECE
6
Mar 5, 2026
0.18Weighted ECE
6
Mar 5, 2026
242Length
6
Feb 26, 2026
19,050Length
6
Feb 26, 2026
240.5Length
6
Feb 26, 2026
5MACE
6
Feb 26, 2026