arXiv AI

Jais 3: Introducing Arabic-centric Agentic AI

Jais 3 is a fleet of six open Arabic‑centric agentic language models ranging from 0.9 B to 375 B parameters, designed to perform multi‑step tasks, call tools, operate terminals, search the web, and manipulate PDFs, spreadsheets, and presentations while communicating in Arabic. The models are adapted from the K2 Horizon fleet through continual pretraining, supervised fine‑tuning, token preference optimization, and reinforcement learning with verifiable rewards, and they support Modern Standard Arabic and major regional dialects with a 512K‑token context window (except the 0.9 B model). Evaluations show that across sizes Jais 3 matches or outperforms comparable open‑weight models on Arabic and agentic benchmarks, with strong instruction following and fewer hallucinations, and the authors release all weights, checkpoints, training data, and evaluation code.

arXiv AI
Aug 17

Jais 2: A Family of Arabic-Centric Open Large Language Models

arXiv:2608. 13580v1 Announce Type: cross Abstract: Jais 2 is a family of Arabic-centric large language models developed jointly by MBZUAI, Cerebras, and Inception, designed to advance Arabic-centric language modeling, with strong performance across the Arabic and culturally grounded benchmarks evaluated in this report.

By Mohamed Anwar, Abed Alhakim Freihat, George Ibrahim, Mostafa Awad, Abdelrahman Sadallah, Gurpreet Gosal, Gokulakrishnan Ramakrishnan, Sarath Chandran, Biswajit Mishra, Rituraj Joshi, Ahmed Frikha, Etienne Goffinet, Abhishek Maiti, Ali El Filali, Sarah AlBarri, Samujjwal Ghosh, Rahul Pal, Parvez Mullah, Awantika Shukla, Sajid siddiki, Samta Kamboj, Onkar Pandit, Sunil Kumar Sahu, AbdelRahman Elbadawy, Amr Mohamed, Ahmad Chamma, Evan Dufraisse, Abdelaziz Bounhar, Dani Bouch, Hadi Abdine, Guokan Shang, Fajri Koto, Yuxia Wang, Zhuohan Xie, Ali Mekky, Rania Elbadry, Sarfraz Ahmad, Momina Ahsan, Omar El Herraoui, Daniil Orel, Hasan Iqbal, Kareem Elzeky, Mervat Abassy, Kareem Elozeiri, Saadeldine Eletter, Farah Atif, Nurdaulet Mukhituly, Haonan Li, Xudong Han, Aaryamonvikram Singh, Zainul Abedien Ahmed Quraishi, Neha Sengupta, Larry Murray, Avraham Sheinin, Joel Hestness, Natalia Vassilieva, Hector Xuguang Ren, Zhengzhong Liu, Michalis Vazirgiannis, Preslav Nakov
arXiv Machine Learning
Sep 11

E-CONAN (Entailment, CONtradition And Neutral) Benchmarks: Arabic Textual Entailment and Natural Inference Datasets

E-CONAN introduces Arabic textual entailment and natural inference benchmarks comprising two datasets: E-CONAN-2 (2-way RTE) and E-CONAN-3 (3-way NLI). The datasets are built from automatically-translated pairs, human-validated machine translations, hand-crafted pairs from Arabic teaching books, and rumor-containing news headlines. The authors evaluated nine multilingual pretrained models and five large language models on these benchmarks, demonstrating that E-CONAN offers a more diverse and robust assessment than existing datasets like XNLI and ArNLI.

By Khloud AL Jallad, Nada Ghneim, Ghaida Rebdawi
arXiv Computation and Language
Sep 11

Nuha-Speech: Building General-Purpose Arabic Speech-LLMs

Nuha‑Speech is a new initiative aimed at creating general‑purpose Arabic speech‑large language models (speech‑LLMs). It includes the construction of a large Arabic Speech Question‑Answering corpus with over 1.5 million samples for instruction tuning, supervised fine‑tuning of Qwen‑Omni model variants at various scales, and a systematic evaluation framework with diverse tasks and tailored metrics. The project seeks to establish foundational infrastructure for Arabic speech‑LLMs amid limited Arabic speech resources.

By Yingzhi Wang, Reem Alhazzani, Muhammad Alqurishi
arXiv Machine Learning
Oct 23

AQL-Lite: Enabling Efficient Knowledge Acquisition for Arabic LLMs via Data-Centric Continual Pretraining

AQL‑Lite is a framework that enables efficient adaptation of multilingual large language models to Modern Standard Arabic by combining a quality‑first data curation pipeline with a partial‑parameter continual pretraining recipe. The approach trains RS‑LoRA adapters, the last eight transformer blocks, and the embedding matrix using layer‑wise learning rates, achieving significant benchmark gains while reducing training time and resource usage. In large‑scale experiments, AQL‑Lite recovers most of the performance of full‑parameter continual pretraining and outperforms adapter‑only baselines, demonstrating its practicality for Arabic LLM development.

By Houssam Eddine Boukhalfa, Mourad Bouache, Mourad Dahmane
arXiv Computation and Language
Aug 28

Why Current XAI Is Not Enough for Arabic NLP: A Critical Survey of the Explainability Gap

The paper surveys the state of Explainable AI (XAI) in Arabic NLP, highlighting three gaps: a method gap where Arabic XAI relies mainly on limited post‑hoc techniques; a task gap with most work focused on classification tasks and little on generation, retrieval, or dialogue; and a linguistic gap where explanations rarely address Arabic‑specific phenomena such as morphology, dialects, and diglossia. It proposes a taxonomy of tasks, methods, linguistic units, and evaluation practices, and outlines a research agenda for linguistically grounded Arabic XAI.

By Salima Lamsiyah, Ruslan Mitkov
Hugging Face Trending Papers
Jul 22

HalluTruthQA: A Fine-Grained Benchmark for Hallucination Detection, Localization, and Explanation in Arabic Question Answering

Large language models (LLMs) can generate fluent Arabic answers, yet factual errors remain difficult to detect, localize, explain, and verify. Existing hallucination benchmarks often provide response-level labels, with limited support for identifying the exact erroneous content, explaining why it is incorrect, or selecting the correct factual answer.

arXiv Computation and Language
Nov 4

Current state of LLMs for Arabic dialectal machine translation

This study evaluates large language models (LLMs) for English‑to‑dialectal Arabic machine translation using the MADAR dataset across 16 Arabic dialects, covering both translation directions. It compares specialized Arabic models (Jais, Nile), multilingual models (Gemma, Command‑R, Mistral, Aya), and commercial APIs (GPT‑4.1) with multiple metrics—BLEU, CHRF, COMET, GEMBA—and a small manual assessment. The authors also examine LLMs’ ability to classify dialects and discuss challenges in automatic MT evaluation for Arabic dialects, noting potential training data contamination.

By Josef Jon, Rawan Bondok, Ond\v{r}ej Bojar
arXiv Computation and Language
Sep 22

AlexandriaX 2026: The First Shared Task on Dialectal Arabic Machine Translation

arXiv:2609.22796v1 Announce Type: new Abstract: Dialectal Arabic machine translation (MT) remains challenging despite recent progress in Arabic language technologies, particularly because effective t...

By Abdellah El Mekki, AbdelRahim A. Elmadany, Samar M. Magdy, Saad Ezzini, Mo El-Haj, Mustafa Jarrar, Zaid Alyafeai, Bernard Ghanem, Muhammad Abdul-Mageed
arXiv AI
3d ago

Evaluating Jev on Arabic: Comparing Decision Models and LLMs Across Arabic Tasks and Dialects

The paper evaluates the Jev system on Arabic, comparing it with 66 other models across 43 tasks and 9 task families. Jev 1.14.0 scores a mean task-primary metric of 69.24, slightly below Gemma-4-31B’s 71.81 but above Jev 1.13.0’s 68.77. The study also examines performance differences between Modern Standard Arabic and dialects, noting variable gaps and a notable Palestinian banking penalty for Jev 1.14.0.

By Hossam Elsafty, Tobias Deu{\ss}er, Lorenz Sparrenberg, Rajkumar Ramamurthy, Rafet Sifa
arXiv Computation and Language
Sep 23

ARAFA: An LLM-Generated Arabic Fact-Checking Dataset

A new large-scale Arabic fact‑checking dataset called Arafa has been created using an automated pipeline that generates claims from Arabic Wikipedia, mutates them into counterfactuals, and validates them against supporting or refuting evidence. The dataset contains 181,976 claim‑evidence pairs labeled as supported, refuted, or not enough information, and human evaluation shows high inter‑annotator agreement and strong validation accuracy. Fine‑tuned transformer models on Arafa achieve a Macro F1‑score of 77%, demonstrating its usefulness for Arabic fact‑checking tasks.

By Christophe Khalil, Shady Elbassuoni, Rida Assaf