# Self-Explaining Hate Speech Detection with Moral Rationales

arXiv: https://arxiv.org/abs/2601.03481 · HTML: https://mlsanity.com/paper/2601.03481/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 74 references that are in the map.
- First posted on arXiv 7 Jan 2026.

## Builds on (most cited first, 22 of 22)

- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [Language (Technology) is Power: A Critical Survey of "Bias" in NLP](https://mlsanity.com/paper/2005.14050/) (2020)
- [ERASER: A Benchmark to Evaluate Rationalized NLP Models](https://mlsanity.com/paper/1911.03429/) (2019)
- [HateXplain: A Benchmark Dataset for Explainable Hate Speech Detection](https://mlsanity.com/paper/2012.10289/) (2020)
- [Nuanced Metrics for Measuring Unintended Bias with Real Data for Text Classification](https://mlsanity.com/paper/1903.04561/) (2019)
- [Predicting the Type and Target of Offensive Posts in Social Media](https://mlsanity.com/paper/1902.09666/) (2019)
- [Annotators with Attitudes: How Annotator Beliefs And Identities Bias Toxic Language Detection](https://mlsanity.com/paper/2111.07997/) (2021)
- [Racial Bias in Hate Speech and Abusive Language Detection Datasets](https://mlsanity.com/paper/1905.12516/) (2019)
- [On Releasing Annotator-Level Labels and Information in Datasets](https://mlsanity.com/paper/2110.05699/) (2021)
- [Contextualizing Hate Speech Classifiers with Post-hoc Explanation](https://mlsanity.com/paper/2005.02439/) (2020)
- [CLIcK: A Benchmark Dataset of Cultural and Linguistic Intelligence in Korean](https://mlsanity.com/paper/2403.06412/) (2024)
- [Entropy-based Attention Regularization Frees Unintended Bias Mitigation from Lists](https://mlsanity.com/paper/2203.09192/) (2022)
- [Disentangling Perceptions of Offensiveness: Cultural and Moral Correlates](https://mlsanity.com/paper/2312.06861/) (2023)
- [From Languages to Geographies: Towards Evaluating Cultural Bias in Hate Speech Datasets](https://mlsanity.com/paper/2404.17874/) (2024)
- [Cultural Compass: Predicting Transfer Learning Success in Offensive Language Detection with Cultural Features](https://mlsanity.com/paper/2310.06458/) (2023)
- [Towards Interpretable Hate Speech Detection using Large Language Model-extracted Rationales](https://mlsanity.com/paper/2403.12403/) (2024)
- ["So You Think You're Funny?": Rating the Humour Quotient in Standup Comedy](https://mlsanity.com/paper/2110.12765/) (2021)
- [Explainable Abuse Detection as Intent Classification and Slot Filling](https://mlsanity.com/paper/2210.02659/) (2022)
- [MFTCXplain: A Multilingual Benchmark Dataset for Evaluating the Moral Reasoning of LLMs through Multi-hop Hate Speech Explanation](https://mlsanity.com/paper/2506.19073/) (2025)
- [Contextual-Lexicon Approach for Abusive Language Detection](https://mlsanity.com/paper/2104.12265/) (2021)
- [Adaptable Moral Stances of Large Language Models on Sexist Content: Implications for Society and Gender Discourse](https://mlsanity.com/paper/2410.00175/) (2024)
- [Aligning Attention with Human Rationales for Self-Explaining Hate Speech Detection](https://mlsanity.com/paper/2511.07065/) (2025)

## Built on by (most cited first, 5 of 5)

- [KVpop -- Key-Value Cache Compression with Predictive Online Pruning](https://mlsanity.com/paper/2607.05061/) (2026)
- [Explain the Flag: Contextualizing Hate Speech Beyond Censorship](https://mlsanity.com/paper/2604.14970/) (2026)
- [Beyond Topical Similarity: Contrastive Evidence Retrieval with Interpretable Attention Alignment in RAG](https://mlsanity.com/paper/2606.01482/) (2026)
- [Less Is Moral: A CHARMing Framework for Moral Foundations Detection in Endorsement Behaviour](https://mlsanity.com/paper/2609.03330/) (2026)
- [FineWeb-CLaR: Culture, Language, and Region Annotations for Benchmark-Aligned Corpus Auditing](https://mlsanity.com/paper/2609.25298/) (2026)

## Abstract (arXiv)

Existing hate speech detection models are often opaque and rely on surface-level lexical cues, which makes them vulnerable to spurious correlations and limits robustness, interpretability and cultural contextualization. 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. 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 while enhancing both faithful and plausible explanations. Although explanations become more concise, sufficiency decreases, indicating more compact and informative rationales. Fairness remains stable, suggesting that improvements in explanation quality do not introduce significant bias trade-offs.