# Socio-Culturally Aware Evaluation Framework for LLM-Based Content Moderation

arXiv: https://arxiv.org/abs/2412.13578 · HTML: https://mlsanity.com/paper/2412.13578/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Dec 2024.

## Built on by (most cited first, 5 of 5)

- [Evaluating Large Language Models for Detecting Antisemitism](https://mlsanity.com/paper/2509.18293/) (2025)
- [A Framework for Situating Innovations, Opportunities, and Challenges in Advancing Vertical Systems with Large AI Models](https://mlsanity.com/paper/2504.02793/) (2025)
- [CRaFT: An Explanation-Based Framework for Evaluating Cultural Reasoning in Multilingual Language Models](https://mlsanity.com/paper/2510.14014/) (2025)
- [WHEN TO ACT, WHEN TO WAIT: Modeling the Intent-Action Alignment Problem in Dialogue](https://mlsanity.com/paper/2506.01881/) (2025)
- [Mod-Guide: An LLM-based Content Moderation Feedback System to Address Insensitive Speech toward Indigenous Ethnic and Religious Minority Communities](https://mlsanity.com/paper/2606.13397/) (2026)

## Abstract (arXiv)

With the growth of social media and large language models, content moderation has become crucial. Many existing datasets lack adequate representation of different groups, resulting in unreliable assessments. To tackle this, we propose a socio-culturally aware evaluation framework for LLM-driven content moderation and introduce a scalable method for creating diverse datasets using persona-based generation. Our analysis reveals that these datasets provide broader perspectives and pose greater challenges for LLMs than diversity-focused generation methods without personas. This challenge is especially pronounced in smaller LLMs, emphasizing the difficulties they encounter in moderating such diverse content.