HateBRXplain: A Benchmark Dataset with Human-Annotated Rationales for Explainable Hate Speech Detection in Brazilian Portuguese

Isadora Salles, Francielle Vargas, Fabrício Benevenuto


Abstract
Nowadays, hate speech technologies are surely relevant in Brazil. Nevertheless, the inability of these technologies to provide reasons (rationales) for their decisions is the limiting factor to their adoption since they comprise bias, which may perpetuate social inequalities when propagated at scale. This scenario highlights the urgency of proposing explainable technologies to address hate speech. However, explainable models heavily depend on data availability with human-annotated rationales, which are scarce, especially for low-resource languages. To fill this relevant gap, we introduce HateBRXplain, the first benchmark dataset for hate speech detection in Portuguese, with text span annotations capturing rationales. We evaluated our corpus using mBERT, BERTimbau, DistilBERTimbau, and PTT5 models, which outperformed the current baselines. We further assessed these models’ explainability using model-agnostic explanation methods (LIME and SHAP). Results demonstrate plausible post-hoc explanations when compared to human annotations. However, the best-performing hate speech detection models failed to provide faithful rationales.
Anthology ID:
2025.coling-main.446
Volume:
Proceedings of the 31st International Conference on Computational Linguistics
Month:
January
Year:
2025
Address:
Abu Dhabi, UAE
Editors:
Owen Rambow, Leo Wanner, Marianna Apidianaki, Hend Al-Khalifa, Barbara Di Eugenio, Steven Schockaert
Venue:
COLING
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
6659–6669
Language:
URL:
https://aclanthology.org/2025.coling-main.446/
DOI:
Bibkey:
Cite (ACL):
Isadora Salles, Francielle Vargas, and Fabrício Benevenuto. 2025. HateBRXplain: A Benchmark Dataset with Human-Annotated Rationales for Explainable Hate Speech Detection in Brazilian Portuguese. In Proceedings of the 31st International Conference on Computational Linguistics, pages 6659–6669, Abu Dhabi, UAE. Association for Computational Linguistics.
Cite (Informal):
HateBRXplain: A Benchmark Dataset with Human-Annotated Rationales for Explainable Hate Speech Detection in Brazilian Portuguese (Salles et al., COLING 2025)
Copy Citation:
PDF:
https://aclanthology.org/2025.coling-main.446.pdf