Mod-Guide: An LLM-based Content Moderation Feedback System to Address Insensitive Speech toward Indigenous Ethnic and Religious Minority Communities

The paper presents Mod-Guide, an LLM-based content moderation system designed to address culturally insensitive speech towards Indigenous and religious minority communities in Bangladesh, specifically targeting the Hindu and Chakma populations. It utilizes a retrieval augmented generation (RAG) approach, incorporating a culturally grounded corpus co-created with community members to enhance the sensitivity of moderation responses. This advancement is significant for practitioners as it highlights the importance of integrating minority perspectives into AI systems, improving contextual accuracy and ethical considerations in content moderation.