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 4 · SOTA2 Research

Natural Language Inference

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

73.37Accuracy

21

Feb 26, 2026

81.2Avg Accuracy

21

Feb 26, 2026

ContractNLI re-annotated (test)

83Accuracy

20

Jun 16, 2026

0.03Loss

20

Jun 1, 2026

MNLI GLUE (test)

86.56Accuracy

20

May 29, 2026

EntailmentBank (test)

llama3-70b-8192

54BLEU

20

Feb 26, 2026

33.1Accuracy

20

Feb 26, 2026

91.45MNLI Accuracy

20

May 14, 2026

QNLI 64 instances (test)

91.1Accuracy

20

Feb 26, 2026

94.6Accuracy

19

Feb 26, 2026

SNLI 1.0 (test)

BERT-base w/ Z-Aug

90.67Accuracy

19

Feb 26, 2026

93.15MNLI Acc

18

Feb 26, 2026

59.1Accuracy

18

Apr 23, 2026

Ro-BERT + Cart-Stra-CL++

75Micro F1 Score

18

Feb 26, 2026

65.34Accuracy

18

Feb 26, 2026

65.68Accuracy

18

Feb 26, 2026

71.65Accuracy

18

Feb 26, 2026

NLI adversarial benchmark (test)

BERT-base w/ Seq-Z

75.4Average Score

18

Feb 26, 2026

MultiNLI matched (in-domain) RepEval 2017 (test)

76.8Accuracy

18

Feb 26, 2026

CMNLI syntactically perturbed

76.67Accuracy

17

Apr 2, 2026

68.15Accuracy

17

Apr 2, 2026

95.78Accuracy

17

Mar 24, 2026

ALUM_ROBERTA-LARGE-SMART

98.2Accuracy

17

Feb 26, 2026

2.08ECE

16

Jun 5, 2026

English NLI (dev)

62.73Accuracy

16

Apr 28, 2026

Page 4 of 15

Popular datasets

MNLI ANLI SNLI XNLI MultiNLI

Follow for updates

Join SOTA2 Contact