In this paper we investigate the explainability of transformer models and their plausibility for hate speech and counter speech detection. We compare representatives of four different explainability approaches, i.e., gradient-based, perturbation-based, attention-based, and prototype-based approaches, and analyze them quantitatively with an ablation study and qualitatively in a user study. Results show that perturbation-based explainability performs best, followed by gradient-based and attention-based explainability. Prototypebased experiments did not yield useful results. Overall, we observe that explainability strongly supports the users in better understanding the model predictions.
@article{arxiv.2407.20274,
title = {Exploring the Plausibility of Hate and Counter Speech Detectors with Explainable AI},
author = {Adrian Jaques Böck and Djordje Slijepčević and Matthias Zeppelzauer},
journal= {arXiv preprint arXiv:2407.20274},
year = {2024}
}