# Sadeed: Advancing Arabic Diacritization Through Small Language Model

arXiv: https://arxiv.org/abs/2504.21635 · HTML: https://mlsanity.com/paper/2504.21635/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 28 references that are in the map.
- First posted on arXiv 30 Apr 2025.

## Builds on (most cited first, 9 of 9)

- [Gemma 2: Improving Open Language Models at a Practical Size](https://mlsanity.com/paper/2408.00118/) (2024)
- [Aya 23: Open Weight Releases to Further Multilingual Progress](https://mlsanity.com/paper/2405.15032/) (2024)
- [ALLaM: Large Language Models for Arabic and English](https://mlsanity.com/paper/2407.15390/) (2024)
- [Neural Arabic Text Diacritization: State of the Art Results and a Novel Approach for Machine Translation](https://mlsanity.com/paper/1911.03531/) (2019)
- [Deep Diacritization: Efficient Hierarchical Recurrence for Improved Arabic Diacritization](https://mlsanity.com/paper/2011.00538/) (2020)
- [Arabic Diacritic Recovery Using a Feature-Rich biLSTM Model](https://mlsanity.com/paper/2002.01207/) (2020)
- [Arabic Text Diacritization In The Age Of Transfer Learning: Token Classification Is All You Need](https://mlsanity.com/paper/2401.04848/) (2024)
- [Kuwain 1.5B: An Arabic SLM via Language Injection](https://mlsanity.com/paper/2504.15120/) (2025)
- [Take the Hint: Improving Arabic Diacritization with Partially-Diacritized Text](https://mlsanity.com/paper/2306.03557/) (2023)

## Abstract (arXiv)

Arabic text diacritization remains a persistent challenge in natural language processing due to the language's morphological richness. In this paper, we introduce Sadeed, a novel approach based on a fine-tuned decoder-only language model adapted from Kuwain 1.5B Hennara et al. [2025], a compact model originally trained on diverse Arabic corpora. Sadeed is fine-tuned on carefully curated, high-quality diacritized datasets, constructed through a rigorous data-cleaning and normalization pipeline. Despite utilizing modest computational resources, Sadeed achieves competitive results compared to proprietary large language models and outperforms traditional models trained on similar domains. Additionally, we highlight key limitations in current benchmarking practices for Arabic diacritization. To address these issues, we introduce SadeedDiac-25, a new benchmark designed to enable fairer and more comprehensive evaluation across diverse text genres and complexity levels. Together, Sadeed and SadeedDiac-25 provide a robust foundation for advancing Arabic NLP applications, including machine translation, text-to-speech, and language learning tools.