Metraipolis › archive › dc_afbeee2b3665
Training-Time Explainability for Multilingual Hate Speech Detection: Aligning Model Reasoning with Human Rationales — arXiv:2608.26125
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.26125
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T14:23:59.193Z
- SHA-256 of the archived text
1107e8451f3ddd9a2b19113295fa98706e31278729c6a254a8c40cae4e3069a8- Stable URL of this page
https://metraipolis.com/documento/dc_afbeee2b3665
How to cite
Training-Time Explainability for Multilingual Hate Speech Detection: Aligning Model Reasoning with Human Rationales — arXiv:2608.26125 — arXiv (en), https://arxiv.org/abs/2608.26125 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_afbeee2b3665 (Metraipolis archive dc_afbeee2b3665, sha256 1107e8451f3ddd9a2b19113295fa98706e31278729c6a254a8c40cae4e3069a8)
BibTeX
@misc{dcafbeee2b3665,
title = {Training-Time Explainability for Multilingual Hate Speech Detection: Aligning Model Reasoning with Human Rationales — arXiv:2608.26125},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.26125},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_afbeee2b3665; sha256 1107e8451f3ddd9a2b19113295fa98706e31278729c6a254a8c40cae4e3069a8},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.26125. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Online hate against Muslim communities often appears in culturally coded, multilingual forms that evade conventional AI moderation. Such systems, though accurate, remain opaque and risk bias, over-censorship, or under-moderation, particularly when detached from sociocultural context. We propose a \emph{training-time} explainability framework that aligns model reasoning with human-annotated rationales, improving both classification performance and interpretability. Our approach is evaluated on HateXplain (English) and BullySent (Hinglish), reflecting the prevalence of anti-Muslim hate across both languages. Using LIME, Integrated Gradients, Grad X Input, and attention, we assess accuracy, explanation quality, and cross-method agreement. Results show that gradient- and attention-based regularization improve F-scores, enhance plausibility and faithfulness, and capture culturally specific cues for detecting implicit anti-Muslim hate, offering a path toward multilingual, culturally aware content moderation.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-28 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.26125 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T13:16:53.584Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).