arXiv Open Access 2025

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

Saeed Almheiri Rania Hossam Mena Attia Chenxi Wang Preslav Nakov +2 lainnya
Lihat Sumber

Abstrak

Large language models (LLMs) often reflect Western-centric biases, limiting their effectiveness in diverse cultural contexts. Although some work has explored cultural alignment, the potential for cross-cultural transfer, using alignment in one culture to improve performance in others, remains underexplored. This paper investigates 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 13 Arab countries, we evaluate lightweight alignment methods such as in-context learning and demonstration-based reinforcement (DITTO), alongside baselines like supervised fine-tuning and direct preference optimization. Our results show that merely 12 culture-specific examples from one country can improve performance in others by 10\% on average, within multilingual models. In addition, we demonstrate that out-of-culture demonstrations from Indonesia and US contexts can match or surpass in-culture alignment for MCQ reasoning, 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.

Topik & Kata Kunci

Penulis (7)

S

Saeed Almheiri

R

Rania Hossam

M

Mena Attia

C

Chenxi Wang

P

Preslav Nakov

T

Timothy Baldwin

F

Fajri Koto

Format Sitasi

Almheiri, S., Hossam, R., Attia, M., Wang, C., Nakov, P., Baldwin, T. et al. (2025). Cross-Cultural Transfer of Commonsense Reasoning in LLMs: Evidence from the Arab World. https://arxiv.org/abs/2509.19265

Akses Cepat

Lihat di Sumber
Informasi Jurnal
Tahun Terbit
2025
Bahasa
en
Sumber Database
arXiv
Akses
Open Access ✓