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

Winogender dataset and benchmarks · SOTA2 Research

Research Datasets

Winogender

Benchmarks

Task NameDataset NameSOTA MethodSort

Task NameDataset NameSOTA ResultTrendResults

Attention Head Selection (Causal Tracing)

WinoGender

0.008Loss (10% Sparsity)

24

Coreference Resolution

Winogender

64.72Accuracy

15

Coreference Resolution

Winogender (test)

80.7Accuracy

11

Coreference Resolution

Winogender (WG) (test)

80.8Accuracy

11

Winogender

65.3Normalized Log Accuracy

9

Bias Evaluation

WinoGender

0.068EBS

8

Commonsense Reasoning

WinoGender

0.968Accuracy

8

Multiple-choice scoring

Winogender

0.847Accuracy

7

Co-reference resolution

WinoGender

77.5Accuracy (All)

4

Pronoun Coreference Resolution

Winogender (test)

64.5Accuracy

3

Popular tasks

Coreference Resolution Commonsense Reasoning Bias Evaluation Safety Multiple-choice scoring

Follow for updates

Join SOTA2 Contact