EMNLP Findings2025
MAKIEval: A Multilingual Automatic WiKidata-based Framework for Cultural Awareness Evaluation for LLMs
Raoyuan Zhao, Beiduo Chen, Barbara Plank, Michael A. Hedderich
Abstract
Large language models (LLMs) are used globally across many languages, but their English-centric pretraining raises concerns about cross-lingual disparities for cultural awareness, often resulting in biased outputs. However, comprehensive multilingual evaluation remains challenging due to limited benchmarks and questionable translation quality. To better assess these disparities, we introduce MAKIEval, an automatic multilingual framework for evaluating cultural awareness in LLMs across languages, regions, and topics. MAKIEval evaluates open-ended text generation, capturing how models express culturally grounded knowledge in natural language. Leveraging Wikidata's multilingual structure as a cross-lingual anchor, it automatically identifies cultural entities in model outputs and links them to structured knowledge, enabling scalable, language-agnostic evaluation without manual annotation or translation. We then introduce four metrics that capture complementary dimensions of cultural awareness: granularity, diversity, cultural specificity, and consensus across languages. We assess 7 LLMs developed from different parts of the world, encompassing both open-source and proprietary systems, across 13 languages, 19 countries and regions, and 6 culturally salient topics (e.g., food, clothing). Notably, we find that models tend to exhibit stronger cultural awareness in English, suggesting that English prompts more effectively activate culturally grounded knowledge.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
1 top-tier paper has cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
Cited by top-tier papers1
Builds on15
- Toxicity in chatgpt: Analyzing persona-assigned language modelsAmeet Deshpande, Vishvak Murahari, Tanmay Rajpurohit, Ashwin Kalyan et al.EMNLP Findings 2023 · 178 citations
- Global MMLU: Understanding and Addressing Cultural and Linguistic Biases in Multilingual EvaluationShivalika Singh, Angelika Romanou, Clémentine Fourrier, David Ifeoluwa Adelani et al.ACL 2025 · 144 citations
- "Kelly is a Warm Person, Joseph is a Role Model": Gender Biases in LLM-Generated Reference LettersYixin Wan, George Pu, Jiao Sun, Aparna Garimella et al.EMNLP Findings 2023 · 139 citations
- Marked Personas: Using Natural Language Prompts to Measure Stereotypes in Language ModelsMyra Cheng, Esin Durmus, Dan JurafskyACL 2023 · 89 citations
- Benchmarking Vision Language Models for Cultural UnderstandingShravan Nayak, Kanishk Jain, Rabiul Awal, Siva Reddy et al.EMNLP 2024 · 26 citations
Related papers
- Having Beer after Prayer? Measuring Cultural Bias in Large Language ModelsTarek Naous, Michael J. Ryan, Alan Ritter, Wei XuACL 2024
- MMAC: A Multilingual, Multimodal Alignment Framework for Cultural Grounding EvaluationWeihua Zheng, Zhengyuan Liu, Tanmoy Chakraborty, Weiwen Xu et al.ACL 2026
- XLQA: A Benchmark for Locale-Aware Multilingual Open-Domain Question AnsweringKeon-Woo Roh, Yeong-Joon Ju, Seong-Whan LeeEMNLP 2025
- Culture In a Frame: C3B as a Comic-Based Benchmark for Multimodal Culturally AwarenessYuchen Song, Andong Chen, Wenxin Zhu, Kehai Chen et al.ICLR 2026 · 3 citations
- GIMMICK: Globally Inclusive Multimodal Multitask Cultural Knowledge BenchmarkingFlorian Schneider, Carolin Holtermann, Chris Biemann, Anne LauscherACL Findings 2025 · 4 citations
