# AdaCultureSafe: Adaptive Cultural Safety Grounded by Cultural Knowledge in Large Language Models

arXiv: https://arxiv.org/abs/2603.08275 · HTML: https://mlsanity.com/paper/2603.08275/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 40 of its 65 references that are in the map.
- First posted on arXiv 9 Mar 2026.

## Builds on (most cited first, 30 of 40)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [A Survey on LLM-as-a-Judge](https://mlsanity.com/paper/2411.15594/) (2024)
- [Layer by Layer: Uncovering Hidden Representations in Language Models](https://mlsanity.com/paper/2502.02013/) (2025)
- [Think-on-Graph: Deep and Responsible Reasoning of Large Language Model on Knowledge Graph](https://mlsanity.com/paper/2307.07697/) (2023)
- [AlphaEdit: Null-Space Constrained Knowledge Editing for Language Models](https://mlsanity.com/paper/2410.02355/) (2024)
- [Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications](https://mlsanity.com/paper/2402.05162/) (2024)
- [BLEnD: A Benchmark for LLMs on Everyday Knowledge in Diverse Cultures and Languages](https://mlsanity.com/paper/2406.09948/) (2024)
- [CVQA: Culturally-diverse Multilingual Visual Question Answering Benchmark](https://mlsanity.com/paper/2406.05967/) (2024)
- [Graph Chain-of-Thought: Augmenting Large Language Models by Reasoning on Graphs](https://mlsanity.com/paper/2404.07103/) (2024)
- [Generative Language Models Exhibit Social Identity Biases](https://mlsanity.com/paper/2310.15819/) (2023)
- [Survey of Cultural Awareness in Language Models: Text and Beyond](https://mlsanity.com/paper/2411.00860/) (2024)
- [Large Language Models are Geographically Biased](https://mlsanity.com/paper/2402.02680/) (2024)
- [Neurons in Large Language Models: Dead, N-gram, Positional](https://mlsanity.com/paper/2309.04827/) (2023)
- [SafetyPrompts: a Systematic Review of Open Datasets for Evaluating and Improving Large Language Model Safety](https://mlsanity.com/paper/2404.05399/) (2024)
- [AceGPT, Localizing Large Language Models in Arabic](https://mlsanity.com/paper/2309.12053/) (2023)
- [CLIcK: A Benchmark Dataset of Cultural and Linguistic Intelligence in Korean](https://mlsanity.com/paper/2403.06412/) (2024)
- [Massively Multi-Cultural Knowledge Acquisition & LM Benchmarking](https://mlsanity.com/paper/2402.09369/) (2024)
- [Graph-constrained Reasoning: Faithful Reasoning on Knowledge Graphs with Large Language Models](https://mlsanity.com/paper/2410.13080/) (2024)
- [Evaluating Cultural and Social Awareness of LLM Web Agents](https://mlsanity.com/paper/2410.23252/) (2024)
- [An Evaluation of Cultural Value Alignment in LLM](https://mlsanity.com/paper/2504.08863/) (2025)
- [CultureGuard: Towards Culturally-Aware Dataset and Guard Model for Multilingual Safety Applications](https://mlsanity.com/paper/2508.01710/) (2025)
- [NormDial: A Comparable Bilingual Synthetic Dialog Dataset for Modeling Social Norm Adherence and Violation](https://mlsanity.com/paper/2310.14563/) (2023)
- [BertaQA: How Much Do Language Models Know About Local Culture?](https://mlsanity.com/paper/2406.07302/) (2024)
- [Cultural Commonsense Knowledge for Intercultural Dialogues](https://mlsanity.com/paper/2402.10689/) (2024)
- [Arabic Dataset for LLM Safeguard Evaluation](https://mlsanity.com/paper/2410.17040/) (2024)
- [Palm: A Culturally Inclusive and Linguistically Diverse Dataset for Arabic LLMs](https://mlsanity.com/paper/2503.00151/) (2025)
- [Demystifying the Roles of LLM Layers in Retrieval, Knowledge, and Reasoning](https://mlsanity.com/paper/2510.02091/) (2025)
- [GraphTool-Instruction: Revolutionizing Graph Reasoning in LLMs through Decomposed Subtask Instruction](https://mlsanity.com/paper/2412.12152/) (2024)
- [NileChat: Towards Linguistically Diverse and Culturally Aware LLMs for Local Communities](https://mlsanity.com/paper/2505.18383/) (2025)

## Abstract (arXiv)

With the global proliferation of Large Language Models (LLMs), cultural safety, defined as the ability to generate respectful and appropriate responses across diverse cultures, becomes critical for responsible AI deployment. However, existing research often treats cultural safety and cultural knowledge in isolation. It remains unclear whether cultural safety is grounded in understanding varying cultural knowledge to enable LLMs to adaptively yield respectful and appropriate responses across diverse cultures, which trigger ethical concerns in cross-cultural scenarios. In this work, we introduce AdaCultureSafe, a dataset designed to jointly evaluate cultural safety and knowledge. Through AdaCultureSafe, we reveal a critical insight: textita significant decoupling exists between cultural safety and cultural knowledge proficiency. Although LLMs possess rich cultural knowledge, they fail to leverage it to improve cultural safety. LLMs tend to rely on generic safety rather than safety based on culture-specific knowledge. Motivated by this, we propose a knowledge-grounded method that elicits internal cultural knowledge in LLMs during response generation. Experimental results demonstrate that our approach effectively improves cultural safety. Our work can aid better understanding the landscape of cultural safety of LLMs.