# FanarGuard: A Culturally-Aware Moderation Filter for Arabic Language Models

arXiv: https://arxiv.org/abs/2511.18852 · HTML: https://mlsanity.com/paper/2511.18852/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 55 references that are in the map.
- First posted on arXiv 24 Nov 2025.

## Builds on (most cited first, 24 of 24)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [Jailbroken: How Does LLM Safety Training Fail?](https://mlsanity.com/paper/2307.02483/) (2023)
- [HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal](https://mlsanity.com/paper/2402.04249/) (2024)
- [A General Theoretical Paradigm to Understand Learning from Human Preferences](https://mlsanity.com/paper/2310.12036/) (2023)
- [Enhancing Chat Language Models by Scaling High-quality Instructional Conversations](https://mlsanity.com/paper/2305.14233/) (2023)
- [Aligning AI With Shared Human Values](https://mlsanity.com/paper/2008.02275/) (2020)
- [WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs](https://mlsanity.com/paper/2406.18495/) (2024)
- [Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks](https://mlsanity.com/paper/2404.02151/) (2024)
- [Safety-Tuned LLaMAs: Lessons From Improving the Safety of Large Language Models that Follow Instructions](https://mlsanity.com/paper/2309.07875/) (2023)
- [PKU-SafeRLHF: Towards Multi-Level Safety Alignment for LLMs with Human Preference](https://mlsanity.com/paper/2406.15513/) (2024)
- [SALAD-Bench: A Hierarchical and Comprehensive Safety Benchmark for Large Language Models](https://mlsanity.com/paper/2402.05044/) (2024)
- [Towards Measuring and Modeling "Culture" in LLMs: A Survey](https://mlsanity.com/paper/2403.15412/) (2024)
- [AceGPT, Localizing Large Language Models in Arabic](https://mlsanity.com/paper/2309.12053/) (2023)
- [PolyGuard: A Multilingual Safety Moderation Tool for 17 Languages](https://mlsanity.com/paper/2504.04377/) (2025)
- [From Instructions to Intrinsic Human Values -- A Survey of Alignment Goals for Big Models](https://mlsanity.com/paper/2308.12014/) (2023)
- [What's the Magic Word? A Control Theory of LLM Prompting](https://mlsanity.com/paper/2310.04444/) (2023)
- [Culture is Not Trivia: Sociocultural Theory for Cultural NLP](https://mlsanity.com/paper/2502.12057/) (2025)
- [Command R7B Arabic: A Small, Enterprise Focused, Multilingual, and Culturally Aware Arabic LLM](https://mlsanity.com/paper/2503.14603/) (2025)
- [Palm: A Culturally Inclusive and Linguistically Diverse Dataset for Arabic LLMs](https://mlsanity.com/paper/2503.00151/) (2025)
- [X-Guard: Multilingual Guard Agent for Content Moderation](https://mlsanity.com/paper/2504.08848/) (2025)
- [Decolonial AI Alignment: Openness, Vi'sedsa-Dharma, and Including Excluded Knowledges](https://mlsanity.com/paper/2309.05030/) (2023)
- [Explaining the role of Intrinsic Dimensionality in Adversarial Training](https://mlsanity.com/paper/2405.17130/) (2024)

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

- [IndicGuard: A Multilingual Safety Guard Model and Dataset for Indic Languages](https://mlsanity.com/paper/2606.22841/) (2026)
- [ArabicDialectSafety: A Dialect-Aware Benchmark for Arabic Content Safety Classification](https://mlsanity.com/paper/2608.01291/) (2026)
- [Beyond Cultural Knowledge: Evaluating Arabic Cultural Appropriateness of Large Language Models](https://mlsanity.com/paper/2609.16006/) (2026)
- [ArGuard Shared Task: Harmful Content Detection in Arabic Memes and LLM Prompts](https://mlsanity.com/paper/2609.29349/) (2026)
- [From Normative Frameworks to Alignment Data: Constructing and Evaluating SFT and Preference Data](https://mlsanity.com/paper/2609.35201/) (2026)

## Abstract (arXiv)

Content moderation filters are a critical safeguard against alignment failures in language models. Yet most existing filters focus narrowly on general safety and overlook cultural context. In this work, we introduce FanarGuard, a bilingual moderation filter that evaluates both safety and cultural alignment in Arabic and English. We construct a dataset of over 468K prompt and response pairs, drawn from synthetic and public datasets, scored by a panel of LLM judges on harmlessness and cultural awareness, and use it to train two filter variants. To rigorously evaluate cultural alignment, we further develop the first benchmark targeting Arabic cultural contexts, comprising over 1k norm-sensitive prompts with LLM-generated responses annotated by human raters. Results show that FanarGuard achieves stronger agreement with human annotations than inter-annotator reliability, while matching the performance of state-of-the-art filters on safety benchmarks. These findings highlight the importance of integrating cultural awareness into moderation and establish FanarGuard as a practical step toward more context-sensitive safeguards.