Preprint · arXiv · October 18, 2026
Aligning Cultural Behaviors in Arabic Language Models via Preferences
Junhong Liang, Daniil Orel, Saeed Almheiri, Bilal Elbouardi, Younes Samih, Abed Alhakim Freihat
Abstract
Arabic LLMs have improved rapidly, but aligning them with country-specific cultural behavioral preferences remains underexplored.Previous work has focused primarily on instructiontuning for factual knowledge, overlooking nuanced cultural behaviors.We propose ARABPREF, which to our knowledge is the first large-scale pan-Arab preference data set for the alignment of cultural behavior conditioned on the country.Our contribution includes two resources in English and Modern Standard Arabic: a culturally grounded preference dataset for training and evaluation, and a multiple-choice benchmark designed to assess culturally aligned behavior across nations.All test data are validated by native speakers to ensure authenticity.To optimize cultural alignment, we experiment with post-training methods across multilingual and Arabic-centric language models, evaluating performance on both generation tasks and our cultural multiplechoice benchmark.By releasing both a training dataset and an evaluation benchmark, we provide a foundation for systematically studying and improving culturally aligned behavior in language models.
More by Daniil Orel
- Jais 2: A Family of Arabic-Centric Open Large Language ModelsarXiv · 2026 · Citations: 3
- Stereotype Bias in a Bilingual Setting: A Culturally Grounded Evaluation in KazakhstanUnderline Science Inc. · 2026
- Beyond Memorization: Extending Reasoning Depth with Recurrence, Memory and Test-Time Compute ScalingUnderline Science Inc. · 2026
- A Multi-View Media Profiling Suite: Resources, Evaluation, and AnalysisUnderline Science Inc. · 2026
