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

2026-06-11

AI Summary

This study addresses the limitations of current large language models in detecting subtle cultural offenses—such as cultural erasure and misrepresentation—targeting Indigenous and religious minority groups during content moderation. Focusing on Hindu and Chakma communities in Bangladesh, the work introduces hermeneutic inclusivity and restorative justice principles into moderation system design for the first time. Through a community-engaged approach, the authors construct a culturally sensitive corpus and integrate minority narratives into the moderation pipeline using retrieval-augmented generation (RAG). Mixed-methods evaluation demonstrates that the RAG-enhanced system significantly improves contextual accuracy in culturally nuanced scenarios, while feedback from diverse ethnic groups reveals differential perceptions of its outputs. These findings validate that incorporating marginalized communities’ lived experiences effectively enhances model sensitivity to cultural context.

Abstract

Language operates as a mechanism of both marginalization and resistance, especially for minority communities navigating insensitive and harmful speech online. As content moderation increasingly depends on large language models (LLMs), concerns arise about whether these systems can recognize culturally insensitive speech-language that disregards or marginalizes the cultural and religious perspectives of historically underrepresented communities, often through implicit erasure, misrepresentation, or normative framing, rather than overt hostility. Focusing on Bangladesh's Hindu and Chakma communities -- the country's largest religious and Indigenous ethnic minorities, respectively -- this paper investigates the epistemic limits of LLM-based moderation systems and explores methods for incorporating minority perspectives. We co-created a culturally grounded corpus of insensitive speech with community members and integrated their narratives into moderation pipelines using retrieval augmented generation (RAG). Our tool, Mod-Guide, improves LLM sensitivity to minority viewpoints by leveraging contextual cues derived from lived experience. Through mixed-method evaluations involving both minority and majority participants, we demonstrate that RAG-enhanced moderation responses are more contextually accurate and perceived differently across ethnic lines. This work advances research in human-computer interaction, AI ethics, and social computing by foregrounding restorative justice and hermeneutical inclusion in the design of content moderation systems.

Problem

• content moderation
• insensitive speech
• minority communities
• large language models
• cultural sensitivity

Innovation

• retrieval-augmented generation
• culturally grounded corpus
• content moderation
• minority perspectives
• hermeneutical inclusion