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

Multiple-Choice Classification research benchmarks · SOTA2 Research

Multiple-Choice Classification

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

Llama-Guard-3-8B

100Accuracy

47

Jun 2, 2026

Task Calibration

78.63Accuracy

24

May 12, 2026

83.4Accuracy

16

Feb 26, 2026

66.9Accuracy

16

Feb 26, 2026

89.6Accuracy

16

Feb 26, 2026

84.5Accuracy

16

Feb 26, 2026

OpenBookQA ARC-Easy WinoGrande HellaSwag PIQA MathQA

35Accuracy (OpenBookQA)

11

May 26, 2026

Standard Benchmarks (ARC-Challenge, HellaSwag, Lambada, PIQA, WinoGrande) Zero-Shot

56.31ARC-C Accuracy

9

Jun 26, 2026

14 standard NLP tasks suite (held-out)

86.9StoryCloze

8

Feb 26, 2026

Popular datasets

MMLU StrategyQA AGNews E-SNLI NLP tasks

Follow for updates

Join SOTA2 Contact