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

Hate Speech Detection research benchmarks – Page 3 · SOTA2 Research

Hate Speech Detection

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

Hate-esBas Spanish (test)

74.91Macro F1

5

Feb 26, 2026

HateBos Italian (test)

74.36Macro F1

5

Feb 26, 2026

Twitter Offensive

Full dataset tuning

84.5Accuracy

5

Feb 26, 2026

HatefulMemes (dev)

73.4ROC AUC

5

Feb 26, 2026

Stormfront -> GHC (target set)

52.9Target F1 Score

5

Feb 26, 2026

HatEval -> GHC (target set)

51.5Target F1

5

Feb 26, 2026

HatEval Spanish

0.542Macro F1

4

Feb 26, 2026

HateSpeech (test)

79.1ASR

4

Feb 26, 2026

79.81ASR

4

Feb 26, 2026

-265.86ΔPPL

4

Feb 26, 2026

Hateful-Memes (HM) (val)

modality relation distillation paradigm

69.85Accuracy

2

May 8, 2026

Sudan v2 (test)

datavaluepeople/Hate-Speech-Sudan-v2

0.92Accuracy

1

Apr 24, 2026

Page 3 of 3

Popular datasets

HatEval HateXplain ETHOS LT-EDI LGBTQ+ Hate Speech Detection Meme Dataset GHC

Follow for updates

Join SOTA2 Contact