Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Uncertainty Calibration research benchmarks – Page 2 · SOTA2 Research

Uncertainty Calibration

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

0.05MACE

6

Feb 26, 2026

0.08MACE

6

Feb 26, 2026

0.059MACE

6

Feb 26, 2026

0.2MACE

6

Feb 26, 2026

Clinical dataset (test)

Uncertainty-Aware Foundation Model

0.041ECE

5

Apr 7, 2026

Beijing Air Quality (test)

0.307Calibration AUC

5

Feb 26, 2026

GSM8k OOD (test)

Entropy-based (ours)

3.15ECE

4

Mar 9, 2026

TriviaQA + Natural Questions ID held-out (test)

Entropy-based (ours)

7.2ECE

4

Mar 9, 2026

0.009ECE

4

Feb 26, 2026

0.059ECE

4

Feb 26, 2026

0.013ECE

4

Feb 26, 2026

Clinc150 (test)

0.021ECE

4

Feb 26, 2026

Banking77 (test)

0.039ECE

4

Feb 26, 2026

3.3ECE

4

Feb 26, 2026

5.1ECE

4

Feb 26, 2026

2.2ECE

4

Feb 26, 2026

AMSR2 (Ice Pack)

Bayesian Transformer

0.003Weighted ECE

3

Mar 5, 2026

Bayesian Transformer

0.19Weighted ECE

3

Mar 5, 2026

Places365 OOD (test)

48.65RMS Calibration Error

3

Feb 26, 2026

Texture OOD (test)

28.74RMS Calibration Error

3

Feb 26, 2026

SVHN OOD (test)

45.32RMS Calibration Error

3

Feb 26, 2026

MNIST OOD (test)

0.513RMS Calibration Error

3

Feb 26, 2026

Tiny ImageNet OOD (test)

31.99RMS Calibration Error

3

Feb 26, 2026

CIFAR100 OOD (test)

33.04RMS Calibration Error

3

Feb 26, 2026

WCMAnatGridRes1004

0.023alpha_T

2

May 4, 2026

Page 2 of 3

Popular datasets

CIFAR ImageNet meps GPQA Sentinel-1

Follow for updates

Join SOTA2 Contact