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

Long-context Language Understanding research benchmarks – Page 4 · SOTA2 Research

Long-context Language Understanding

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

LongBench Zh out-of-domain (evaluation)

Original Prompt

61.2SingleDoc Acc

3

Feb 26, 2026

LongBench English (14-task average)

11.6LongBench Average Score

2

May 14, 2026

57.96HotpotQA

2

May 11, 2026

LongBench-E 128K context

MiniCPM3-4B (MLA)

43.28Average Score

2

Apr 15, 2026

Page 4 of 4

Popular datasets

LongBench RULER InfiniteBench SCROLLS L-Eval

Follow for updates

Join SOTA2 Contact