Elamur Research⌕
☰
home / papers / 2025 / The Landscape of Arabic Large Language Models (ALLMs): A New Era for Arabic Language Technology

The Landscape of Arabic Large Language Models (ALLMs): A New Era for Arabic Language Technology

Summary

This survey charts how Arabic language technology grew from early rule-based tools into today's Arabic large language models (ALLMs). Arabic is hard for NLP because its morphology is rich, its dialects are many, and there is little digital text in it. The authors trace that history in phases: morphological analyzers from the 1980s, then statistical n-gram and stemming methods, then embeddings and neural toolkits, then BERT-style encoders, and finally generative and multimodal ALLMs that followed ChatGPT. They also describe how these models are built. Pretraining corpora are filtered and de-duplicated. Instruction data is mostly translated from English. Evaluation has moved from single-task test sets to suites that probe reasoning, culture, dialects and specific domains. The survey's main conclusion is that machine translation made billion-parameter Arabic models feasible, but it also brought in Western cultural bias and left dialects underrepresented. The paper closes with an agenda: benchmarks that are authentic and culturally grounded, scalable human evaluation, a pan-Arab data consortium, and better links from research to deployment.

Key points

  • According to the paper, Arabic is the native language of more than 422 million people across 27 countries, and Arabic NLP began with Sakhr Software in 1985.
  • Toolkits and encoders covered:
    • toolkits: MADAMIRA, FARASA, CAMeL;
    • encoders: AraBERT, CAMEL-BERT, QARIB;
    • generative models: JASMINE, Jais/Jais-chat, Atlas-Chat, AceGPT, ALLaM, Fanar, Peacock, Dallah.
  • Training pipelines differ by model:
    • Jais filters pretraining data with rules, while Fanar uses syntactic, semantic and model-based filters plus in-house MT (English→MSA, MSA→dialects);
    • ALLaM's work reports that training without translated data gave higher loss.
  • Instruction data is mostly translated from English sets (Super-NaturalInstructions, Natural Questions, P3, xP3). Jais adds in-house sets: NativeQA (regional facts) plus SafetyQA and DoNotAnswer for refusals. CIDER targets cultural alignment.
  • Benchmarks evolved over time:
    • early and broad suites: AraBench, then ALUE, ARLUE, ARGEN;
    • LLM-era sets: ORCA, Dolphin, AlGhafa, Qiyas, ArabicMMLU, AraSWAG, ARCD, AraTruthfulQA, ACVA, ArabLegalEval, CAMEL-Bench;
    • AraDICE covers MSA, Egyptian and Gulf Arabic via post-edited MT;
    • Fanar was tested by over 300 people from many Arab countries.

Interesting ideas

  • I1. Building a dialect benchmark by machine-translating items and then having humans post-edit them (AraDICE) gets broad coverage cheaply while keeping quality checked.
  • I2. Fanar filters data on several levels (syntactic, semantic and model-based) rather than with rules alone, so each layer catches different kinds of noise.
  • I3. Jais pairs ordinary QA data with sets of questions the model should refuse (SafetyQA, DoNotAnswer), so the model learns both what to answer and what to decline.
  • I4. ALLaM compares two model responses side by side, and Fanar uses a large, regionally diverse pool of human testers; both give richer judgments than single-score metrics.
  • I5. The survey treats translated training data as a trade-off: it enables scale and lowers loss but imports cultural bias, so it uses separate cultural benchmarks (ACVA) to measure the side effect.

For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

  • I1 (answer): Fable/tale QA sets could be drafted automatically and then hand-corrected, which grows test coverage without full manual authoring.
  • I2 (learn): Silver trees agreed by the witness parsers could also pass syntactic validity checks and frame-level semantic consistency checks before entering training.
  • I3 (answer): Adding items the reader should mark as unanswerable from the text works as a negative control that exposes guessing.
  • I4 (back): Generated or reconstructed texts from the text→graph→text compressor could be judged by pairwise preference against the original or rival outputs.
  • I5 (world): Knowledge compiled from translated or foreign-centric sources could be tagged by origin and checked with a dedicated probe set, so source bias is visible rather than hidden.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Linguistics & languages

Topics Language models & pretraining Semitic languages

Cite

Cite as Shahad Al-Khalifa; Nadir Durrani; Hend Al-Khalifa; and Firoj Alam. 2025. The Landscape of Arabic Large Language Models (ALLMs): A New Era for Arabic Language Technology. arXiv:2506.01340.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.2506.01340 · arXiv 2506.01340

@misc{alkhalifa2025landscape,
  title = {{The Landscape of Arabic Large Language Models (ALLMs): A New Era for Arabic Language Technology}},
  author = {Shahad Al-Khalifa; Nadir Durrani; Hend Al-Khalifa; and Firoj Alam},
  year = {2025},
  doi = {10.48550/arxiv.2506.01340},
  eprint = {2506.01340},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2506.01340}
}

References

31 references, 15 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Ahmed Abdelali, Kareem Darwish, Nadir Durrani, and Hamdy Mubarak. 2016. Farasa: A Fast and Furious Segmenter for Arabic. In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations, John DeNero, Mark Finlayson, and Sravana Reddy (Eds.). Association for Computational Linguistics, San Diego, California, 11–16. https://doi.org/10.18653/v1/N16-3003 ↗ our card
  2. Ahmed Abdelali, Sabit Hassan, Hamdy Mubarak, Kareem Darwish, and Younes Samih. 2021. Pre-Training BERT on Arabic Tweets: Practical Considerations. (2021). arXiv:2102.10684 [cs.CL] ↗ our card
  3. Muhammad Abdul-Mageed, AbdelRahim Elmadany, and El Moatez Billah Nagoudi. 2021. ARBERT & MARBERT: Deep Bidirectional Transformers for Arabic. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), Chengqing Zong, Fei Xia, Wenjie Li, and Roberto Navigli (Eds.). Association for Computational Linguistics, Online, 7088–7105. https://doi.org/10.18653/v1/2021.acl-long.551 ↗ our card
  4. Shahad Al-Khalifa and Hend Al-Khalifa. 2024. The Qiyas Benchmark: Measuring ChatGPT Mathematical and Language Understanding in Arabic. In Proceedings of the 7th International Conference on Natural Language and Speech Processing (ICNLSP 2024), Mourad Abbas and Abed Alhakim Freihat (Eds.). Association for Computational Linguistics, Trento, 343–351. https://aclanthology.org/2024.icnlsp-1.35/ ↗
  5. Muhammad Asif Ali, Nawal Daftardar, Mutayyba Waheed, Jianbin Qin, and Di Wang. 2025. MQA-KEAL: Multi-hop Question Answering under Knowledge Editing for Arabic Language. In Proceedings of the 31st International Conference on Computational Linguistics, Owen Rambow, Leo Wanner, Marianna Apidianaki, Hend Al-Khalifa, Barbara Di Eugenio, and Steven Schockaert (Eds.). Association for Computational Linguistics, Abu Dhabi, UAE, 5629–5644. https://aclanthology.org/2025.coling-main.377/ ↗
  6. Ebtesam Almazrouei, Ruxandra Cojocaru, Michele Baldo, Quentin Malartic, Hamza Alobeidli, Daniele Mazzotta, Guilherme Penedo, Giulia Campesan, Mugariya Farooq, Maitha Alhammadi, Julien Launay, and Badreddine Noune. 2023. AlGhafa Evaluation Benchmark for Arabic Language Models. In Proceedings of ArabicNLP 2023, Hassan Sawaf, Samhaa El-Beltagy, Wajdi Zaghouani, Walid Magdy, Ahmed Abdelali, Nadi Tomeh, Ibrahim Abu Farha, Nizar Habash, Salam Khalifa, Amr Keleg, Hatem Haddad, Imed Zitouni, Khalil Mrini, and Rawan Almatham (Eds.). Association for Computational Linguistics, Singapore (Hybrid), 244–275. https://doi.org/10.18653/v1/2023.arabicnlp-1.21 ↗
  7. Fakhraddin Alwajih, Gagan Bhatia, and Muhammad Abdul-Mageed. 2024. Dallah: A Dialect-Aware Multimodal Large Language Model for Arabic. In Proceedings of The Second Arabic Natural Language Processing Conference, Nizar Habash, Houda Bouamor, Ramy Eskander, Nadi Tomeh, Ibrahim Abu Farha, Ahmed Abdelali, Samia Touileb, Injy Hamed, Yaser Onaizan, Bashar Alhafni, Wissam Antoun, Salam Khalifa, Hatem Haddad, Imed Zitouni, Badr AlKhamissi, Rawan Almatham, and Khalil Mrini (Eds.). Association for Computational Linguistics, Bangkok, Thailand, 320–336. https://doi.org/10.18653/v1/2024.arabicnlp-1.27 ↗
  8. Fakhraddin Alwajih, El Moatez Billah Nagoudi, Gagan Bhatia, Abdelrahman Mohamed, and Muhammad Abdul-Mageed. 2024. Peacock: A Family of Arabic Multimodal Large Language Models and Benchmarks. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Lun-Wei Ku, Andre Martins, and Vivek Srikumar (Eds.). Association for Computational Linguistics, Bangkok, Thailand, 12753–12776. https://doi.org/10.18653/v1/2024.acl-long.689 ↗
  9. Zaid Alyafeai, Khalid Almubarak, Ahmed Ashraf, Deema Alnuhait, Saied Alshahrani, Gubran Abdulrahman, Gamil Ahmed, Qais Gawah, Zead Saleh, Mustafa Ghaleb, Yousef Ali, and Maged Al-shaibani. 2024. CIDAR: Culturally Relevant Instruction Dataset For Arabic. In Findings of the Association for Computational Linguistics: ACL 2024, Lun-Wei Ku, Andre Martins, and Vivek Srikumar (Eds.). Association for Computational Linguistics, Bangkok, Thailand, 12878–12901. https://doi.org/10.18653/v1/2024.findings-acl.764 ↗
  10. Wissam Antoun, Fady Baly, and Hazem Hajj. 2020. AraBERT: Transformer-based Model for Arabic Language Understanding. In Proceedings of the 4th Workshop on Open-Source Arabic Corpora and Processing Tools, with a Shared Task on Offensive Language Detection, Hend Al-Khalifa, Walid Magdy, Kareem Darwish, Tamer Elsayed, and Hamdy Mubarak (Eds.). European Language Resource Association, Marseille, France, 9–15. https://aclanthology.org/2020.osact-1.2/ ↗ our card
  11. M Saiful Bari, Yazeed Alnumay, Norah A. Alzahrani, Nouf M. Alotaibi, Hisham Abdullah Alyahya, Sultan AlRashed, Faisal Abdulrahman Mirza, Shaykhah Z. Alsubaie, Hassan A. Alahmed, Ghadah Alabduljabbar, Raghad Alkhathran, Yousef Almushayqih, Raneem Alnajim, Salman Alsubaihi, Maryam Al Mansour, Saad Amin Hassan, Dr. Majed Alrubaian, Ali Alammari, Zaki Alawami, Abdulmohsen Al-Thubaity, Ahmed Abdelali, Jeril Kuriakose, Abdalghani Abujabal, Nora Al-Twairesh, Areeb Alowisheq, and Haidar Khan. 2025. ALLaM: Large Language Models for Arabic and English. In The Thirteenth International Conference on Learning Representations. https://openreview.net/forum?id=MscdsFVZrN ↗ our card
  12. El Moatez Billah Nagoudi, Muhammad Abdul-Mageed, AbdelRahim Elmadany, Alcides Inciarte, and Md Tawkat Islam Khondaker. 2023. JASMINE: Arabic GPT Models for Few-Shot Learning. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, Houda Bouamor, Juan Pino, and Kalika Bali (Eds.). Association for Computational Linguistics, Singapore, 16721–16744. https://doi.org/10.18653/v1/2023.emnlp-main.1040 ↗
  13. Naaima Boudad, Rdouan Faizi, Rachid Oulad Haj Thami, and Raddouane Chiheb. 2018. Sentiment analysis in Arabic: A review of the literature. Ain Shams Engineering Journal 9, 4 (2018), 2479–2490. find
  14. AbdelRahim Elmadany, ElMoatez Billah Nagoudi, and Muhammad Abdul-Mageed. 2023. ORCA: A Challenging Benchmark for Arabic Language Understanding. In Findings of the Association for Computational Linguistics: ACL 2023, Anna Rogers, Jordan Boyd-Graber, and Naoaki Okazaki (Eds.). Association for Computational Linguistics, Toronto, Canada, 9559–9586. https://doi.org/10.18653/v1/2023.findings-acl.609 ↗
  15. Sara Ghaboura, Ahmed Heakl, Omkar Thawakar, Ali Alharthi, Ines Riahi, Abduljalil Saif, Jorma Laaksonen, Fahad S. Khan, Salman Khan, and Rao M. Anwer. 2024. CAMEL-Bench: A Comprehensive Arabic LMM Benchmark. arXiv preprint arXiv:2410.18976 (2024). arXiv:2410.18976 [cs.CV] https://arxiv.org/abs/2410.18976 ↗
  16. Faris Hijazi, Somayah Alharbi, Abdulaziz AlHussein, Harethah Shairah, Reem Alzahrani, Hebah Alshamlan, George Turkiyyah, and Omar Knio. 2024. ArabLegalEval: A Multitask Benchmark for Assessing Arabic Legal Knowledge in Large Language Models. In Proceedings of the Second Arabic Natural Language Processing Conference, Nizar Habash, Houda Bouamor, Ramy Eskander, Nadi Tomeh, Ibrahim Abu Farha, Ahmed Abdelali, Samia Touileb, Injy Hamed, Yaser Onaizan, Bashar Alhafni, Wissam Antoun, Salam Khalifa, Hatem Haddad, Imed Zitouni, Badr AlKhamissi, Rawan Almatham, and Khalil Mrini (Eds.). Association for Computational Linguistics, Bangkok, Thailand, 225–249. https://doi.org/10.18653/v1/2024.arabicnlp-1.20 ↗
  17. Huang Huang, Fei Yu, Jianqing Zhu, Xuening Sun, Hao Cheng, Song Dingjie, Zhihong Chen, Mosen Alharthi, Bang An, Juncai He, Ziche Liu, Junying Chen, Jianquan Li, Benyou Wang, Lian Zhang, Ruoyu Sun, Xiang Wan, Haizhou Li, and Jinchao Xu. 2024. AceGPT, Localizing Large Language Models in Arabic. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), Kevin Duh, Helena Gomez, and Steven Bethard (Eds.). Association for Computational Linguistics, Mexico City, Mexico, 8139–8163. https://doi.org/10.18653/v1/2024.naacl-long.450 ↗ our card
  18. Go Inoue, Bashar Alhafni, Nurpeiis Baimukan, Houda Bouamor, and Nizar Habash. 2021. The Interplay of Variant, Size, and Task Type in Arabic Pre-trained Language Models. In Proceedings of the Sixth Arabic Natural Language Processing Workshop, Nizar Habash, Houda Bouamor, Hazem Hajj, Walid Magdy, Wajdi Zaghouani, Fethi Bougares, Nadi Tomeh, Ibrahim Abu Farha, and Samia Touileb (Eds.). Association for Computational Linguistics, Kyiv, Ukraine (Virtual), 92–104. https://aclanthology.org/2021.wanlp-1.10/ ↗ our card
  19. Malak Mansour, Ahmed Aly, Bahey Tharwat, Sarim Hashmi, Dong An, and Ian Reid. 2025. Language and Planning in Robotic Navigation: A Multilingual Evaluation of State-of-the-Art Models. arXiv preprint arXiv:2501.05478 (2025). ↗ our card
  20. Pedro Henrique Martins, Patrick Fernandes, João Alves, Nuno M. Guerreiro, Ricardo Rei, Duarte M. Alves, José Pombal, Amin Farajian, Manuel Faysse, Mateusz Klimaszewski, Pierre Colombo, Barry Haddow, José G. C. de Souza, Alexandra Birch, and André F. T. Martins. 2024. EuroLLM: Multilingual Language Models for Europe. arXiv:2409.16235 [cs.CL] https://arxiv.org/abs/2409.16235 ↗ our card
  21. Malak Mashaabi, Shahad Al-Khalifa, and Hend Al-Khalifa. 2024. A Survey of Large Language Models for Arabic Language and its Dialects. arXiv preprint arXiv:2410.20238 (2024). ↗ our card
  22. El Moatez Billah Nagoudi, AbdelRahim Elmadany, and Muhammad Abdul-Mageed. 2021. AraT5: Text-to-text transformers for Arabic language generation. arXiv preprint arXiv:2109.12068 (2021). ↗ our card
  23. El Moatez Billah Nagoudi, AbdelRahim Elmadany, Ahmed El-Shangiti, and Muhammad Abdul-Mageed. 2023. Dolphin: A Challenging and Diverse Benchmark for Arabic NLG. In Findings of the Association for Computational Linguistics: EMNLP 2023, Houda Bouamor, Juan Pino, and Kalika Bali (Eds.). Association for Computational Linguistics, Singapore, 1404–1422. https://doi.org/10.18653/v1/2023.findings-emnlp.98 ↗
  24. Xuan-Phi Nguyen, Wenxuan Zhang, Xin Li, Mahani Aljunied, Zhiqiang Hu, Chenhui Shen, Yew Ken Chia, Xingxuan Li, Jianyu Wang, Qingyu Tan, Liying Cheng, Guanzheng Chen, Yue Deng, Sen Yang, Chaoqun Liu, Hang Zhang, and Lidong Bing. 2024. SeaLLMs – Large Language Models for Southeast Asia. arXiv:2312.00738 [cs.CL] https://arxiv.org/abs/2312.00738 ↗
  25. Ossama Obeid, Nasser Zalmout, Salam Khalifa, Dima Taji, Mai Oudah, Bashar Alhafni, Go Inoue, Fadhl Eryani, Alexander Erdmann, and Nizar Habash. 2020. CAMeL tools: An open source python toolkit for Arabic natural language processing. In Proceedings of the twelfth language resources and evaluation conference. 7022–7032. our card
  26. Arfath Pasha, Mohamed Al-Badrashiny, Mona Diab, Ahmed El Kholy, Ramy Eskander, Nizar Habash, Manoj Pooleery, Owen Rambow, and Ryan Roth. 2014. MADAMIRA: A Fast, Comprehensive Tool for Morphological Analysis and Disambiguation of Arabic. In Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC’14), Nicoletta Calzolari, Khalid Choukri, Thierry Declerck, Hrafn Loftsson, Bente Maegaard, Joseph Mariani, Asuncion Moreno, Jan Odijk, and Stelios Piperidis (Eds.). European Language Resources Association (ELRA), Reykjavik, Iceland, 1094–1101. https://aclanthology.org/L14-1479/ ↗ our card
  27. Hassan Sajjad, Ahmed Abdelali, Nadir Durrani, and Fahim Dalvi. 2020. AraBench: Benchmarking Dialectal Arabic-English Machine Translation. In Proceedings of the 28th International Conference on Computational Linguistics. International Committee on Computational Linguistics, Barcelona, Spain (Online), 5094–5107. https://doi.org/10.18653/v1/2020.coling-main.447 ↗
  28. Haitham Seelawi, Ibraheem Tuffaha, Mahmoud Gzawi, Wael Farhan, Bashar Talafha, Riham Badawi, Zyad Sober, Oday Al-Dweik, Abed Alhakim Freihat, and Hussein Al-Natsheh. 2021. ALUE: Arabic Language Understanding Evaluation. In Proceedings of the Sixth Arabic Natural Language Processing Workshop, Nizar Habash, Houda Bouamor, Hazem Hajj, Walid Magdy, Wajdi Zaghouani, Fethi Bougares, Nadi Tomeh, Ibrahim Abu Farha, and Samia Touileb (Eds.). Association for Computational Linguistics, Kyiv, Ukraine (Virtual), 173–184. https://aclanthology.org/2021.wanlp-1.18/ ↗
  29. Neha Sengupta, Sunil Kumar Sahu, Bokang Jia, Satheesh Katipomu, Haonan Li, Fajri Koto, William Marshall, Gurpreet Gosal, Cynthia Liu, Zhiming Chen, et al. 2023. Jais and jais-chat: Arabic-centric foundation and instruction-tuned open generative large language models. arXiv preprint arXiv:2308.16149 (2023). ↗ our card
  30. Abu Bakr Soliman, Kareem Eissa, and Samhaa R. El-Beltagy. 2017. AraVec: A set of Arabic Word Embedding Models for use in Arabic NLP. Procedia Computer Science 117 (2017), 256–265. https://doi.org/10.1016/j.procs.2017.10.117 Arabic Computational Linguistics. ↗
  31. Fanar Team, Ummar Abbas, Mohammad Shahmeer Ahmad, Firoj Alam, Enes Altinisik, Ehsannedin Asgari, Yazan Boshmaf, Sabri Boughorbel, Sanjay Chawla, Shammur Chowdhury, Fahim Dalvi, Kareem Darwish, Nadir Durrani, Mohamed Elfeky, Ahmed Elmagarmid, Mohamed Eltabakh, Masoomali Fatehkia, Anastasios Fragkopoulos, Maram Hasanain, Majd Hawasly, Mus’ab Husaini, Soon-Gyo Jung, Ji Kim Lucas, Walid Magdy, Safa Messaoud, Abubakr Mohamed, Tasnim Mohiuddin, Basel Mousi, Hamdy Mubarak, Ahmad Musleh, Zan Naeem, Mourad Ouzzani, Dorde Popovic, Amin Sadeghi, Husrev Taha Sencar, Mohammed Shinoy, Omar Sinan, Yifan Zhang, Ahmed Ali, Yassine El Kheir, Xiaosong Ma, and Chaoyi Ruan. 2025. Fanar: An Arabic-Centric Multimodal Generative AI Platform. (2025). arXiv:2501.13944 [cs.CL] https://arxiv.org/abs/2501.13944 ↗ our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/al-khalifa-2025-landscape-arabic-large-language-models/