Abstract
Hate speech detection models rely on surface-level lexical features, increasing vulnerability to spurious correlations and limiting robustness, cultural contextualization, and interpretability. We propose Supervised Moral Rationale Attention (SMRA), the first self-explaining hate speech detection framework to incorporate moral rationales as direct supervision for attention alignment. Based on Moral Foundations Theory, SMRA aligns token-level attention with expert-annotated moral rationales, guiding models to attend to morally salient spans rather than spurious lexical patterns. Unlike prior rationale-supervised or post-hoc approaches, SMRA integrates moral rationale supervision directly into the training objective, producing inherently interpretable and contextualized explanations. To support our framework, we also introduce HateBRMoralXplain, a Brazilian Portuguese benchmark dataset annotated with hate labels, moral categories, token-level moral rationales, and socio-political metadata. Across binary hate speech detection and multi-label moral sentiment classification, SMRA consistently improves performance (e.g., +0.9 and +1.5 F1, respectively) while substantially enhancing explanation faithfulness, increasing IoU F1 (+7.4 pp) and Token F1 (+5.0 pp). Although explanations become more concise, sufficiency improves (+2.3 pp) and fairness remains stable, indicating more faithful rationales without performance or bias trade-offs
Abstract (translated)
仇恨言论检测模型依赖于表面级词汇特征，这增加了对虚假相关性的脆弱性，并限制了其鲁棒性、文化语境化和可解释性。我们提出了监督道德理由注意（SMRA），这是一种首个自我解释的仇恨言论检测框架，该框架将道德理由作为注意力对齐的直接监督。基于道德基础理论，SMRA使词级注意力与专家标注的道德理由对齐，引导模型关注道德上重要的短语而非虚假词汇模式。不同于之前的依据理由监管或事后方法，SMRA直接在训练目标中集成了道德理由监督，从而生成固有的可解释和文化背景化的解释。 为了支持我们的框架，我们还引入了HateBRMoralXplain，这是一个巴西葡萄牙语基准数据集，该数据集根据仇恨标签、道德类别、词级道德理由及社会政治元数据进行了标注。在二元仇恨言论检测和多标签道德情绪分类任务中，SMRA显著提高了性能（例如，分别增加了0.9 F1和1.5 F1），同时大幅增强了解释的一致性，IoU F1提高7.4个百分点，Token F1提高5.0个百分点。虽然解释变得更加简洁，但充足度有所增加（+2.3 pp），公平性保持稳定，这表明更忠实的理由并未影响性能或偏见权衡。
URL
https://arxiv.org/abs/2601.03481
