LLMs recognize out-of-vocabulary Islamophobic slurs on platforms like 4Chan and Telegram, which score higher in toxicity than other hate categories, with topics spanning far-right and political movements targeting Muslim immigrants.
Is- lamophobia content detection using natural language processing
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Decoding Islamophobic Discourse: Using LLMs to Identify Tropes and Semi-Coded Hate Speech
LLMs recognize out-of-vocabulary Islamophobic slurs on platforms like 4Chan and Telegram, which score higher in toxicity than other hate categories, with topics spanning far-right and political movements targeting Muslim immigrants.