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 13 · SOTA2 Research

Natural Language Inference

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

Urdu NLI (test)

49.4Accuracy

4

Apr 28, 2026

51.84Accuracy

4

Apr 28, 2026

Hindi NLI (test)

54.4Accuracy

4

Apr 28, 2026

Hindi NLI (dev)

54.49Accuracy

4

Apr 28, 2026

Arabic NLI (test)

54Accuracy

4

Apr 28, 2026

Arabic NLI (dev)

54.08Accuracy

4

Apr 28, 2026

Korean NLI (test)

62.07Accuracy

4

Apr 28, 2026

LogiQA NLI 2.0 (test)

LSS (Llama-3.1-8B)

56.2Accuracy

4

Apr 23, 2026

RTE low-data regime GLUE

58.53Accuracy

4

Apr 21, 2026

62.4Clean Accuracy

4

Apr 21, 2026

91Cumulative Score (CS)

4

Mar 10, 2026

99.4HANS Accuracy

4

Feb 27, 2026

QNLI Non-Biased

90.9Accuracy

4

Feb 26, 2026

94.3Accuracy

4

Feb 26, 2026

SNLI Non-Biased

0.884Performance

4

Feb 26, 2026

92Accuracy

4

Feb 26, 2026

ANLI Round 3 (R3)

44.43Macro F1

4

Feb 26, 2026

MI-based distillation

4.35ECE

4

Feb 26, 2026

90.7Accuracy

4

Feb 26, 2026

73.6Accuracy

4

Feb 26, 2026

79.6Accuracy

4

Feb 26, 2026

MNLI binary classification (test)

78.7ROC AUC

4

Feb 26, 2026

0.754AUC

4

Feb 26, 2026

6 languages Averaged (test)

53.4Accuracy

4

Feb 26, 2026

6 languages All transfers

43Accuracy

4

Feb 26, 2026

Page 13 of 15

Popular datasets

MNLI ANLI SNLI XNLI MultiNLI

Follow for updates

Join SOTA2 Contact