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

Machine Comprehension research benchmarks · SOTA2 Research

Machine Comprehension

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

0.779Accuracy

80

Feb 26, 2026

77.9Accuracy

77

Feb 26, 2026

AS Reader (+BookTest, ensemble)

83.7Accuracy

56

Feb 26, 2026

81.6Accuracy

56

Feb 26, 2026

AS Reader (+BookTest, ensemble)

85.7Accuracy

37

Feb 26, 2026

AS Reader (+BookTest, ensemble)

82.3Accuracy

37

Feb 26, 2026

NSE Adaptive Computation

73.2Named Entities

12

Feb 26, 2026

MCTest 500 (test)

GPT+Strategies (2x)

82Overall Accuracy

12

Feb 26, 2026

MCTest 160 (test)

GPT+Strategies (2x)

81.7Overall Accuracy

11

Feb 26, 2026

TriviaQA Wikipedia Verified (test)

Embedding Refinement

72.8EM

7

Feb 26, 2026

TriviaQA Web Verified (test)

Clark and Gardner

80EM

5

Feb 26, 2026

TriviaQA Wikipedia Full (test)

46.9EM

5

Feb 26, 2026

MMQA 5.2 v1.0 (test)

m-BERT based MMC

0.744F1 (QE-PE)

4

Feb 26, 2026

translated SQuAD QH-PH 5.1 (test)

m-BERT based MMC

51.6Exact Match (EM)

4

Feb 26, 2026

translated SQuAD QE-PH 5.1 (test)

m-BERT based MMC

52.93Exact Match (EM)

4

Feb 26, 2026

Translated SQuAD QH-PH 5.1 (test)

m-BERT based MMC

63.59F1 Score

4

Feb 26, 2026

Translated SQuAD QH-PE 5.1 (test)

m-BERT based MMC

59.19F1 Score

4

Feb 26, 2026

Translated SQuAD QE-PH 5.1 (test)

m-BERT based MMC

64.51F1

4

Feb 26, 2026

Translated SQuAD QE-PE 5.1 (test)

m-BERT based MMC

94.56F1 Score

4

Feb 26, 2026

TriviaQA Web (test)

40.87EM

3

Feb 26, 2026

Popular datasets

TriviaQA CBT Translated SQuAD SQuAD cnn

Follow for updates

Join SOTA2 Contact