Moderating New Waves of Online Hate with Chain-of-Thought Reasoning in Large Language Models

Open in new window