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

Natural Language Inference research benchmarks – Page 2 · SOTA2 Research

Natural Language Inference

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

Fine-tuning (full)

0.898Accuracy

52

Apr 21, 2026

90.25Accuracy

52

Feb 26, 2026

MultiNLI Matched

80.2Accuracy

49

Feb 26, 2026

91.31Accuracy

46

Feb 27, 2026

90.2Acc (m)

44

Feb 26, 2026

47.2Accuracy

42

Jun 2, 2026

45.4Accuracy (WANLI)

42

Jun 2, 2026

84.8Accuracy (0-shot)

42

Feb 27, 2026

97.9Accuracy

40

Feb 26, 2026

XNLI 1.0 (test)

89.7Accuracy (en)

40

May 14, 2026

Natural Language Inference (NLI) (test)

81.17Accuracy

39

Apr 28, 2026

86.2Accuracy

36

Feb 26, 2026

NLI benchmarks (MNLI, QNLI, SNLI, RTE, SICK, SCITAIL) (test)

96.18MNLI Accuracy

35

Jun 4, 2026

CB SuperGLUE (test)

91.43Accuracy

33

Feb 26, 2026

2.74ECE

32

Jun 5, 2026

81.14Accuracy

32

May 14, 2026

79.07Score

31

Feb 26, 2026

90.7Accuracy

30

Feb 27, 2026

Logan IV et al. (2021)

91Average Accuracy

29

Feb 26, 2026

ANLI R3 1.0 (test)

34.9Weighted F1

28

Feb 26, 2026

ANLI R2 1.0 (test)

0.331Weighted F1

28

Feb 26, 2026

ANLI R1 1.0 (test)

Label-Guided Explanation Selection

41.1Weighted F1

28

Feb 26, 2026

84.3Accuracy

28

Feb 26, 2026

92.2Overall Score

28

Feb 26, 2026

ChaosNLI S easy (test)

Model B (antipignistic target)

62.4Top-1 Accuracy

27

Apr 3, 2026

Page 2 of 15

Popular datasets

MNLI ANLI SNLI XNLI MultiNLI

Follow for updates

Join SOTA2 Contact