analiza aiqbench
This research introduces Mod-Guide, a system that enhances LLM content moderation by integrating Retrieval Augmented Generation (RAG) with culturally specific narratives from Indigenous and religious minority communities in Bangladesh. By leveraging lived experience to train models on implicit erasure rather than just overt hostility, the study demonstrates improved contextual accuracy for protecting marginalized groups online. This work is significant as it bridges AI ethics research with practical human-computer interaction solutions for inclusive digital governance.
Mod-Guide: An LLM-based Content Moderation Feedback System to Address Insensitive Speech toward Indigenous Ethnic and Religious Minority Communities
arXiv:2606.13397v1 Announce Type: cross 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.
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.
Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
arXiv:2606.13397 [cs.HC]
(or arXiv:2606.13397v1 [cs.HC] for this version)
https://doi.org/10.48550/arXiv.2606.13397 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
https://doi.org/10.1145/3811242.3819096 Focus to learn more DOI(s) linking to related resources
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 304 cuvinte
Subiecte conexe
Soccer superstar Kylian Mbappé is the latest global name to find himself on the wrong side of Chinese patriotism
Fortune · 28 oct. 2026
Tool-calling retrieval versus vector RAG for a small Greek--English knowledge base: accuracy and robustness to how users type Greek
arXiv · 7 oct. 2026
Lantern: Conflict-Aware Gradient Blending for Physics-Guided Diffusion Models in Calorimeter Simulation
arXiv · 29 iul. 2026
Narrative Flattening: How Post-Training Compresses Thematic, Affective, and Stylistic Variation in LLM Fiction
arXiv · 9 sept. 2026