This survey comprehensively reviews Arabic text summarization, examining state-of-the-art methodologies, commonly used datasets, and evaluation practices. Despite notable progress, the field faces challenges such as fragmented benchmarking, inconsistent metric use, and lacking resources for long-document summarization. We categorize existing summarization methods into traditional, Transformer-based, and hybrid approaches, highlighting their strengths and limitations. We introduce Mukhtasar, a novel dataset supporting short and long summaries across diverse genres to address significant gaps. Additionally, we propose six standardized evaluation splits tailored to distinct summarization goals, promoting reproducibility and fair comparison. To address inconsistencies, we also recommend a consistent reporting protocol using ROUGE-1, ROUGE-2, ROUGE-L, and ROUGE-S. While lexical overlap metrics dominate evaluation practices, we identify the absence of neural-based metrics for Arabic as a significant limitation and call for future development in this area. Our contributions aim to unify evaluation protocols, enrich available resources, and guide the community toward more interpretable and scalable Arabic summarization research.

Automated essay scoring (AES) has become an essential tool in educational assessment. However, applying AES to the Arabic language presents notable challenges, primarily due to the lack of labeled datasets. This data scarcity hampers the development of reliable machine learning models and slows progress in Arabic natural language processing for educational use. While manual annotation by human experts remains the most accurate method for essay evaluation, it is often too costly and time-consuming to create large-scale datasets, especially for low-resource languages like Arabic. In this work, we introduce a human–AI collaborative framework designed to overcome the shortage of scored Arabic essays. Leveraging QAES, a high-quality annotated dataset, our approach uses Large Language Models (LLMs) to generate multidimensional essay evaluations across seven key writing traits: Relevance, Organization, Vocabulary, Style, Development, Mechanics, and Structure. To ensure accuracy and consistency, we design prompting strategies and validation procedures tailored to each trait. This system is then applied to two unannotated Arabic essay datasets: ZAEBUC and QALB. As a result, we introduce ZaQQ, a newly annotated dataset that merges ZAEBUC, QAES, and QALB. Our findings demonstrate that human–AI collaboration can significantly enhance the availability of labeled resources without compromising assessment quality. The proposed framework serves as a scalable and replicable model for addressing data annotation challenges in low-resource languages and supports the broader goal of expanding access to automated educational assessment tools where expert evaluation is limited.

Mohamed Amin, Mahmoud Rady, Mariam Hossam, Sara Gaballa, Eman Samir, Maria Bassem, Nisreen Hisham, Ayman Khalafallah. Proceedings of The Third Arabic Natural Language Processing Conference: Shared Tasks. 2025.

Physiological signals such as electrocardiograms (ECG) and electroencephalograms (EEG) provide complementary insights into human health and cognition, yet multi-modal integration is challenging due to limited multi-modal labeled data, and modality-specific differences . In this work, we adapt the CBraMod encoder for large-scale self-supervised ECG pretraining, introducing a dual-masking strategy to capture intra- and inter-lead dependencies. To overcome the above challenges, we utilize a pre-trained CBraMod encoder for EEG and pre-train a symmetric ECG encoder, equipping each modality with a rich foundational representation. These representations are then fused via simple embedding concatenation, allowing the classification head to learn cross-modal interactions, together enabling effective downstream learning despite limited multi-modal supervision. Evaluated on emotion recognition, our approach achieves near state-of-the-art performance, demonstrating that carefully designed physiological encoders, even with straightforward fusion, substantially improve downstream performance. These results highlight the potential of foundation-model approaches to harness the holistic nature of physiological signals, enabling scalable, label-efficient, and generalizable solutions for healthcare and affective computing.