Fine-tuning QARiB on Arabic offensive tweets reaches 92% macro F1, and a new transformer-based 'hate speech masking' system achieves 0.30 BLEU-1 without comparison to any baseline.
In: Proceedings of the Third Workshop on Abusive Language Online, pp
1 Pith paper cite this work, alongside 185 external citations. Polarity classification is still indexing.
1
Pith paper citing it
185
external citations · OpenAlex
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
REJECT 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Arabic Hate Speech Identification and Masking in Social Media using Deep Learning Models and Pre-trained Models Fine-tuning
Fine-tuning QARiB on Arabic offensive tweets reaches 92% macro F1, and a new transformer-based 'hate speech masking' system achieves 0.30 BLEU-1 without comparison to any baseline.