Mod-Guide: An LLM-based Content Moderation Feedback System to Address Insensitive Speech toward Indigenous Ethnic and Religious Minority Communities
Researchers have developed Mod-Guide, an LLM-based system designed to improve content moderation for sensitive speech targeting minority communities. The system focuses on the Hindu and Chakma communities in Bangladesh, incorporating their lived experiences and culturally specific narratives into moderation pipelines through retrieval-augmented generation (RAG). Evaluations indicate that RAG-enhanced responses are more contextually accurate and perceived differently by various ethnic groups, advancing research in AI ethics and human-computer interaction. AI
IMPACT This research could lead to more equitable and context-aware AI moderation systems for online platforms.