Rewire at TSRC 2022
More about HateCheck
HateCheck helps researchers and practitioners evaluate and improve hate speech detection models by giving targeted diagnostic insights into model performance. With each of 25+ functional tests in HateCheck, you can find specific model weaknesses and then fix them. Earlier this year, we expanded HateCheck’s coverage from English to 10 more languages, with the support of Google Jigsaw. For more details, visit HateCheck.ai!
Start using HateCheck
Download all HateCheck data and research articles below.
And if you have questions or feedback, we’d love to hear from you!