Research questionHow can machine learning learn from contested labels without assuming one objectively correct target?Standard predictive modeling often treats labels as observations of a single correct target. When judgments differ across sources, disagreement may reflect meaningful perspectives rather than simple annotation error.