We describe the Charles University CUNI-AR submission to the WMT26 General Translation shared task for English into Egyptian Arabic. Our pipeline is built around GEMBA (LLM-as-judge quality) estimator that scores translations independently for accuracy and dialect authenticity. We use it throughout our pipeline: to filter training data, to construct preference pairs for DPO, to select best checkpoints, and to pick the best hypothesis per segment at inference time. Our primary translation model is Gemma-4-12B-it finetuned via QLoRA, we also finetune Jais-2-8B-Chat and Aya-expanse-8B. Gemma-4-12B-it provides a high-quality Egyptian Arabic translation, so SFT yields only marginal quality gains in automatic metrics; DPO is able to improve the metrics it optimizes for, but the real gain in translation quality is to be assessed by human evaluation. The final submission selects the best-scoring translation per segment from 31 candidate checkpoints, with the unmodified base model contributing 31\% of the output.

Quranic Arabic poses important challenges for Arabic NLP due to its distinctive morphology, syntax, and orthography. We present MAQAM, a richly annotated Morphosyntactic Arabic Quranic Analysis and Modeling resource covering the entire Quran, providing 22 lexical, inflectional, and clitic features for every word. The resource is constructed through a semi-automatic pipeline combining feature enrichment, morphological synchronization, and expert manual annotation and validation. We further evaluate Quranic morphosyntactic disambiguation under a realistic setting where the input may be represented in either the Uthmani or Imla’i script, and investigate cross-domain transfer between Quranic Arabic and MSA. Our results show that MSA and Quranic resources are complementary for Quranic disambiguation, with the best performance obtained by combining MSA training data, Quran-specific analyzers, and both script representations, whereas for MSA, only integrating Quranic resources provide modest gains. We make MAQAM publicly available.

This work presents an evaluation of large language models (LLMs) for English to dialectal Arabic machine translation on the MADAR dataset. We evaluate both translation directions (English to Arabic and vice-versa) on 16 Arabic dialects. Our experiments cover a diverse set of models, including specialized Arabic models (Jais, Nile), multilingual models (Gemma, Command-R, Mistral, Aya), and commercial APIs (GPT-4.1). We employ multiple evaluation metrics: BLEU, CHRF, COMET (both reference-based and reference-less variants) and GEMBA (LLM-as-a-judge), as well as a small-scale manual evaluation, to assess translation quality. We discuss the challenges of automatic MT evaluation, especially in the context of Arabic dialects. We also evaluate the ability of LLMs to classify the dialect used in a text. The study offers insights into the capabilities and limitations of current LLMs for dialectal Arabic machine translation, particularly highlighting the difficulty of handling dialectal diversity, although the results may be influenced by possible training data contamination, which is always a concern with LLMs.

Proper nouns in Arabic Wikipedia are frequently undiacritized, creating ambiguity in pronunciation and interpretation, especially for transliterated named entities of foreign origin. While transliteration and diacritization have been well-studied separately in Arabic NLP, their intersection remains underexplored. In this paper, we introduce a new manually diacritized dataset of Arabic proper nouns of various origins with their English Wikipedia equivalent glosses, and present the challenges and guidelines we followed to create it. We benchmark GPT-4o on the task of recovering full diacritization given the undiacritized Arabic and English forms, and analyze its performance. Achieving 73% accuracy, our results underscore both the difficulty of the task and the need for improved models and resources. We release our dataset to facilitate further research on Arabic Wikipedia proper noun diacritization.