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

Open-ended scientific discovery research benchmarks · SOTA2 Research

Open-ended scientific discovery

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

0.667Recall

7

Jul 2, 2026

8Recall

7

Jul 2, 2026

40-task evaluation set Overall 2025

4.78Expert Score (ES)

5

Jun 11, 2026

DiscoveryBench open-ended discovery (evaluation set)

4.91ES

5

Jun 11, 2026

BLADE open-ended discovery (evaluation set)

4.8Expert Score

5

Jun 11, 2026

BixBench open-ended discovery (evaluation set)

4.74ES Score

5

Jun 11, 2026

Popular datasets

iNatDisco BIXBench BLADE DiscoveryBench 40-task evaluation set

Follow for updates

Join SOTA2 Contact