Aladdin-FTI @ AMIYA Three Wishes for Arabic NLP: Fidelity, Diglossia, and Multidialectal Generation
Arabic dialects have long been under-represented in Natural Language Processing (NLP) research due to their non-standardization and high variability, which pose challenges for computational modeling. Recent advances in the field, such as Large Language Models (LLMs), offer promising avenues to address this gap by enabling Arabic to be modeled as a pluricentric language rather than a monolithic system. This paper presents Aladdin-FTI, our submission to the AMIYA shared task. The proposed system is designed to both generate and translate dialectal Arabic (DA). Specifically, the model supports text generation in Moroccan, Egyptian, Palestinian, Syrian, and Saudi dialects, as well as bidirect...
| Authors | Jonathan Mutal, Perla Al Almaoui, Simon Hengchen et al. |
|---|---|
| Primary category | cs.CL |
| Published | 18 February 2026 |
| Last updated at source | 18 February 2026 |
Connected in the knowledge graph
- unige-fti/Aladdin-3BModel · cited by
Models and datasets referencing this paper
- unige-fti/Aladdin-3BModel
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.