LLM 1 Can AI detect the direction of harm? Building a model for message moderation on social media platforms Sep 27, 2024