English

Tox-BART: Leveraging Toxicity Attributes for Explanation Generation of Implicit Hate Speech

Computation and Language 2024-06-07 v1

Abstract

Employing language models to generate explanations for an incoming implicit hate post is an active area of research. The explanation is intended to make explicit the underlying stereotype and aid content moderators. The training often combines top-k relevant knowledge graph (KG) tuples to provide world knowledge and improve performance on standard metrics. Interestingly, our study presents conflicting evidence for the role of the quality of KG tuples in generating implicit explanations. Consequently, simpler models incorporating external toxicity signals outperform KG-infused models. Compared to the KG-based setup, we observe a comparable performance for SBIC (LatentHatred) datasets with a performance variation of +0.44 (+0.49), +1.83 (-1.56), and -4.59 (+0.77) in BLEU, ROUGE-L, and BERTScore. Further human evaluation and error analysis reveal that our proposed setup produces more precise explanations than zero-shot GPT-3.5, highlighting the intricate nature of the task.

Keywords

Cite

@article{arxiv.2406.03953,
  title  = {Tox-BART: Leveraging Toxicity Attributes for Explanation Generation of Implicit Hate Speech},
  author = {Neemesh Yadav and Sarah Masud and Vikram Goyal and Vikram Goyal and Md Shad Akhtar and Tanmoy Chakraborty},
  journal= {arXiv preprint arXiv:2406.03953},
  year   = {2024}
}

Comments

17 Pages, 5 Figures, 13 Tables, ACL Findings 2024

R2 v1 2026-06-28T16:55:40.777Z