Lune

EMNLP Findings2025

Cross-Cultural Transfer of Commonsense Reasoning in LLMs: Evidence from the Arab World

Saeed Almheiri, Rania Elbadry, Mena Attia, Chenxi Wang, Preslav Nakov, Timothy Baldwin, Fajri Koto

Abstract

Large language models (LLMs) often reflect Western-centric biases, which limits their effectiveness in diverse cultural contexts. Although some previous work has explored cultural alignment, the potential for cross-cultural transfer, using alignment in one culture to improve performance in others, remains underexplored. Here we investigate cross-cultural transfer of commonsense reasoning in the Arab world, where linguistic and historical similarities coexist with local cultural differences. Using a culturally grounded commonsense reasoning dataset covering thirteen Arab countries, we evaluate lightweight alignment methods such as in-context learning and demonstration-based reinforcement, alongside baselines such as supervised fine-tuning and direct preference optimization. Our results show that merely twelve culture-specific examples from one country can improve performance in others by 10% on average, within multilingual models. We further demonstrate that out-of-culture demonstrations from Indonesian and US contexts can match or surpass in-culture alignment, highlighting cultural commonsense transferability beyond the Arab world. These findings demonstrate that efficient cross-cultural alignment is possible and offer a promising approach to adapt LLMs to low-resource cultural settings.

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.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • CultureLLM: Incorporating Cultural Differences into Large Language Models cites Not All Countries Celebrate Thanksgiving: On the Cultural Dominance in Large Language Models
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on7

  1. CultureLLM: Incorporating Cultural Differences into Large Language ModelsCheng Li, Mengzhuo Chen, Jindong Wang, Sunayana Sitaram et al.NeurIPS 2024 · 101 citations
  2. Investigating Cultural Alignment of Large Language ModelsBadr AlKhamissi, Muhammad N. ElNokrashy, Mai Alkhamissi, Mona T. DiabACL 2024 · 27 citations
  3. Not All Countries Celebrate Thanksgiving: On the Cultural Dominance in Large Language ModelsWenxuan Wang, Wenxiang Jiao, Jingyuan Huang, Ruyi Dai et al.ACL 2024 · 21 citations
  4. Commonsense Reasoning in Arab CultureAbdelrahman Boda Sadallah, Junior Cedric Tonga, Khalid Almubarak, Saeed Almheiri et al.ACL 2025 · 18 citations
  5. The Echoes of Multilinguality: Tracing Cultural Value Shifts during Language Model Fine-tuningRochelle Choenni, Anne Lauscher, Ekaterina ShutovaACL 2024 · 2 citations

Related papers

  1. Evaluating and Improving Cultural Awareness of Reward Models for LLM AlignmentHongbin Zhang, Kehai Chen, Xuefeng Bai, Yang Xiang et al.ICLR 2026 · 4 citations
  2. Mind the Gap in Cultural Alignment: Task-Aware Culture Management for Large Language ModelsBinchi Zhang, Xujiang Zhao, Jundong Li, Haifeng Chen et al.ACL 2026 · 3 citations
  3. Cultural Benchmarking of LLMs in Standard and Dialectal Arabic DialoguesMuhammad Dehan Al Kautsar, Saeed Almheiri, Momina Ahsan, Bilal Elbouardi et al.ACL 2026
  4. CIDAR: Culturally Relevant Instruction Dataset For ArabicZaid Alyafeai, Khalid Almubarak, Ahmed Ashraf, Deema Alnuhait et al.ACL Findings 2024 · 7 citations
  5. Pearl: A Multimodal Culturally-Aware Arabic Instruction DatasetFakhraddin Alwajih, Samar Mohamed Magdy, Abdellah El Mekki, Omer Nacar et al.EMNLP Findings 2025 · 4 citations
Dusk over the sea between two cliffs drawn in fine vertical lines