Academic and research excellence
Three Research Papers by Students and Researchers of the Faculty of Artificial Intelligence Engineering at Syrian Private University Published in Scopus-Indexed CLEF 2026 Proceedings
The published research addresses advanced applications in artificial intelligence, including the extraction of medical concepts from radiographic images, generative plagiarism detection using hybrid retrieval techniques, and the simplification of scientific texts. This achievement reflects the university's growing academic presence in international forums and its ongoing support for student-driven research.
In a new scientific achievement that enhances the international research presence of the Syrian Private University, three scientific research papers co-authored by faculty members and students from the Faculty of Artificial Intelligence Engineering have been published in the proceedings of the international CLEF 2026 conference, indexed within the Scopus database.
The research papers cover three distinct directions in Artificial Intelligence applications:
First: Artificial Intelligence in Medical Imaging
"Frequency-Aware Multi-Head Learning for Long-Tailed Medical Concept Detection in ImageCLEFmed Caption 2026"
Researchers: Bashar Al-Saleh, Eng. Zeinab Baghdadi, and Dr. Riad Sonbol.
The research presents a methodology based on deep learning and the Swin Transformer architecture to extract medical concepts from radiographic images while addressing the challenges of imbalanced medical datasets. The paper achieved outstanding evaluation results within the accompanying ImageCLEFmed workshop.
Second: Generative Plagiarism Detection Using Hybrid Retrieval Techniques
"Hybrid Sparse-Dense Retrieval with Two-Stage Score Fusion for Generative Plagiarism Detection"
Researchers: Farah Al-Qayyim, Tala Al-Khatib, Eng. Raghad Al-Hosni, and Dr. Riad Sonbol.
The paper introduces a system combining lexical and semantic retrieval to identify original source texts regenerated via Large Language Models (LLMs), utilizing score fusion and re-ranking techniques. Results demonstrated the effectiveness of the proposed approach within the accompanying PAN workshop.
Third: Scientific Text Simplification Using Large Language Models
"A Planning-Based Approach to Scientific Text Simplification"
Researchers: Raneem Rabie, Eng. Wessam Al-Sehli, and Dr. Riad Sonbol.
The study presents a framework relying on prior planning to determine the appropriate simplification strategy for scientific texts before generating the simplified phrasing, enhancing readability and content clarity while preserving scientific precision and meaning.