Seminar: Contextualizing Bias in Hate Speech Detection through Annotator Perspectives

Swabha Swayamdipta, assistant professor of computer science and a Gabilan Assistant Professor at USC, presents on the inadequacies of current methods for debiasing hate speech detection, how the subjectivity of this task design leads to debiasing failures, and the origin of bias in toxic language detection. Swayamdipta will demonstrate how annotators’ demographics and beliefs influence their toxicity ratings, and how ignoring such societal context can lead to biased outcomes.