When Does Demographic Information Help? Data and Modeling Regimes for Perspective-Aware Hate Speech Detection

ArXi:2605.27313v1 Announce Type: new graphic information is often used to model annotator perspectives in subjective tasks such as hate speech detection, but its benefit is inconsistent: it improves performance in some settings and behaves as noise in others. This paper asks when graphic features help. We analyze graphic gain as a function of both data split properties and modeling frameworks. For data splits, we measure annotator disagreement, namely how often annotators assign different labels to the same example, along with.