JAWAHER: A Multidialectal Dataset of Arabic Proverbs for LLM Benchmarking
Magdy, Samar M. Kwon, Sang Yun Alwajih, Fakhraddin Abdelfadil, Safaa Shehata, Shady Abdul-Mageed, Muhammad
Magdy, Samar M.
Kwon, Sang Yun
Alwajih, Fakhraddin
Abdelfadil, Safaa
Shehata, Shady
Abdul-Mageed, Muhammad
Supervisor
Department
Machine Learning
Embargo End Date
Type
Conference proceeding
Date
2025
License
Language
English
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
Recent advancements in instruction finetuning, alignment methods such as reinforcement learning from human feedback (RLHF), and optimization techniques like direct preference optimization (DPO), have significantly enhanced the adaptability of large language models (LLMs) to user preferences. However, despite these innovations, many LLMs continue to exhibit biases toward Western, Anglocentric, or American cultures, with performance on English data consistently surpassing that of other languages. This reveals a persistent cultural gap in LLMs, which complicates their ability to accurately process culturally rich and diverse figurative language, such as proverbs. To address this, we introduce Jawaher, a benchmark designed to assess LLMs' capacity to comprehend and interpret Arabic proverbs. Jawaher includes proverbs from various Arabic dialects, along with idiomatic translations and explanations. Through extensive evaluations of both open- and closed-source models, we find that while LLMs can generate idiomatically accurate translations, they struggle with producing culturally nuanced and contextually relevant explanations. These findings highlight the need for ongoing model refinement and dataset expansion to bridge the cultural gap in figurative language processing. Project GitHub page is accessible at: https://github.com/UBC-NLP/jawaher.
Citation
S. M. Magdy, S. Y. Kwon, F. Alwajih, S. T. Abdelfadil, S. Shehata, and M. Abdul-Mageed, “JAWAHER: A Multidialectal Dataset of Arabic Proverbs for LLM Benchmarking,” vol. 1, pp. 12320–12341, Jun. 2025, doi: 10.18653/V1/2025.NAACL-LONG.613
Source
Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)
Conference
2025 Conference of the North American Chapter of the Association for Computational Linguistics-NAACL
Keywords
Visual Question Answering, Vision-Language Reasoning, Object-Level Understanding, Scene Graphs, Multimodal Pretraining, Benchmark Dataset, Compositional Generalization, Cross-Modal Alignment
Subjects
Source
2025 Conference of the North American Chapter of the Association for Computational Linguistics-NAACL
Publisher
Association for Computational Linguistics