A Comparative Benchmark of Specialized Deep Learning Architectures and Fine-Tuned LLMs for Arabic Text Readability

Arabic News Text Classification Using Deep Learning Models with Dynamic N-grams

This study integrates parallel multi-kernel word-level convolutional features into conventional and hybrid deep learning models for Arabic text analysis tasks, providing a systematic within-study assessment of model sensitivity to architecture, preprocessing, and learning-rate selection.

Benchmarking and AI-assisted human-like evaluation of retrieval-augmented generation for Arabic and English documents

An intelligent deep learning for Arabic stemming and morphological classification

A two-stage deep learning pipeline that combines Arabic text stemming and morphological classification within a single deep learning architecture is proposed that demonstrates a 38% improvement over the best traditional stemmer, P-Stemmer.

STAR: instruction tuning for Arabic across tasks, datasets, and models

An in-depth evaluation of instruction tuning for Arabic NLP tasks using three prominent LLMs: LLaMA 3.1-8B, AceGPT-v2-8B, and Qwen3-8B shows that instruction tuning consistently improves performance across most tasks, with notable variations in effectiveness across different tasks and prompts.

Efficient Multilingual Neural Machine Translation via Corpus-Driven Vocabulary Pruning: An English-Arabic Case Study

This paper proposes a general optimization framework that combines a vocabulary pruning method with a targeted fine-tuning protocol for MNMT models, and reduces the vocabulary size from over 128,000 to approximately 10,000 tokens, enabling a 60% memory saving without any loss in performance.

Performance and Computational Cost of Full and Parameter-Efficient Fine Tuning for Arabic Sentiment Classification Across Training Set Sizes

Pre-trained language models are typically adapted to downstream tasks via full fine tuning. However, this entails substantial computational and memory overhead. Parameter-efficient alternatives such as Low-Rank Adaptation (LoRA), Quantized Low-Rank Adaptation (QLoRA), and frozen backbone mitigate these costs, but empirical evidence on how their performance–cost trade-offs change under low labeled data in Arabic remains limited. This paper compares four adaptation strategies: full fine tuning, frozen backbone, LoRA, and QLoRA for Arabic binary sentiment classification on the Hotel Arabic Reviews Dataset, using CAMeLBERT-Mix as the pre-trained encoder. The methods are evaluated under a unified experimental setting at three labeled-data levels: the full training set, 100 samples per class, and 25 samples per class. The evaluation metrics are reported as means and standard deviations across five random seeds. At the full-data level, full fine tuning, LoRA, and QLoRA achieve macro-F1 scores between 0.9569 and 0.9579 and are comparable within seed variability, while the frozen backbone exhibits performance that is approximately ten points lower. LoRA and QLoRA use approximately 35.0% less peak GPU memory than full fine tuning but require longer training times. Under reduced-data conditions, full fine tuning outperforms all other adaptation strategies with the differences being statically significant.