Muhammad AL-Qurishi

Muhammad AL-Qurishi, Sarah AlQaseemi, Riad Soussi

The effectiveness of the BERT model on multiple linguistic tasks has been well documented. On the other hand, its potentials for narrow and specific domains such as Legal, have not been fully explored. In this paper, we examine how BERT can be used in the Arabic legal domain and try customizing this language model for several downstream tasks using several different domain-relevant training and testing datasets to train BERT from scratch. We introduce the AraLegal-BERT, a bidirectional encoder Transformer-based model that have been thoroughly tested and carefully optimized with the goal to amplify the impact of NLP-driven solution concerning jurisprudence, legal documents, and legal practice. We fine-tuned AraLegal-BERT and evaluated it against three BERT variations for Arabic language in three natural languages understanding (NLU) tasks. The results show that the base version of AraLegal-BERT achieve better accuracy than the general and original BERT over the Legal text.

Bilal Abu-Salih, Muhammad AL-Qurishi, Mohammed Alweshah et al.

The incorporation of data analytics in the healthcare industry has made significant progress, driven by the demand for efficient and effective big data analytics solutions. Knowledge graphs (KGs) have proven utility in this arena and are rooted in a number of healthcare applications to furnish better data representation and knowledge inference. However, in conjunction with a lack of a representative KG construction taxonomy, several existing approaches in this designated domain are inadequate and inferior. This paper is the first to provide a comprehensive taxonomy and a bird's eye view of healthcare KG construction. Additionally, a thorough examination of the current state-of-the-art techniques drawn from academic works relevant to various healthcare contexts is carried out. These techniques are critically evaluated in terms of methods used for knowledge extraction, types of the knowledge base and sources, and the incorporated evaluation protocols. Finally, several research findings and existing issues in the literature are reported and discussed, opening horizons for future research in this vibrant area.

Shuyin Ouyang, Zhaozhi Qian, Faroq AL-Tam et al.

Reinforcement Learning (RL) is an important paradigm for aligning Diffusion Language Models (DLMs) toward functional correctness in code generation. However, these models often encounter a ``capability cliff'' on complex tasks, where execution-based semantic rewards become too low to provide a viable learning signal. In this paper, we present a systematic empirical study of RL post-training for diffusion-based code generation along three axes: reward design, hint-conditioned sampling, and task difficulty. We investigate the effectiveness of execution-free rewards as alternatives to traditional unit-test execution, the role of training-time hint-conditioned diffusion sampling in mitigating exploration bottlenecks, and the impact of these design choices varies across tasks with different difficulty levels. Across HumanEval, MBPP, and LiveCodeBench, we find that static checking is the strongest overall standalone execution-free reward in our setting, especially improving DiffuCoder from 53.9 to 67.1 on HumanEval and from 14.9 to 15.5 on LiveCodeBench while reducing rollout time by 9.4\%. We further find that moderate AST-based hinting is most useful on harder benchmarks, while the best reward design depends strongly on task difficulty: similarity-based rewards are more effective on easier subsets, whereas static checking is more reliable on harder subsets where execution rewards are low. These findings suggest that reward design and training guidance substantially affect diffusion RL performance in our evaluated code-generation setting.

Ignacio Iacobacci, Zhaozhi Qian, Faroq AL-Tam et al.

Recently, a new wave of thinking-capable Large Language Models has emerged, demonstrating exceptional capabilities across a wide range of reasoning benchmarks. Early studies have begun to explore how the amount of compute in terms of the length of the reasoning process, the so-called thinking budget, impacts model performance. In this work, we propose a systematic investigation of the thinking budget as a key parameter, examining its interaction with various configurations such as self-consistency, reflection, and others. Our goal is to provide an informative, balanced comparison framework that considers both performance outcomes and computational cost. Among our findings, we discovered that simply increasing the thinking budget is not the most effective use of compute. More accurate responses can instead be achieved through alternative configurations, such as self-consistency and self-reflection.

Muhammad AL-Qurishi

Given the number of Arabic speakers worldwide and the notably large amount of content in the web today in some fields such as law, medicine, or even news, documents of considerable length are produced regularly. Classifying those documents using traditional learning models is often impractical since extended length of the documents increases computational requirements to an unsustainable level. Thus, it is necessary to customize these models specifically for long textual documents. In this paper we propose two simple but effective models to classify long length Arabic documents. We also fine-tune two different models-namely, Longformer and RoBERT, for the same task and compare their results to our models. Both of our models outperform the Longformer and RoBERT in this task over two different datasets.