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

Instruction Generalization research benchmarks · SOTA2 Research

Instruction Generalization

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

RoleBench Instruction Generalization

57.6CUS Score

10

Feb 26, 2026

RoleBench Chinese instruction generalization 1.0

53.7ROUGE-L (CUS)

7

Feb 26, 2026

OOD Instructions Novel Prompts

95T3 Success Count [C]

6

Apr 28, 2026

RoleBench instruction generalization

55.8GPT-4 Win Rate

5

Feb 26, 2026

Popular datasets

RoleBench OOD Instructions

Follow for updates

Join SOTA2 Contact