Mohamed Adel | Searcharxiv

arXiv2026

Arabic Morphosyntactic Tagging and Dependency Parsing with Large Language Models

LLMs perform strongly across NLP, but their ability to produce explicit grammatical analyses remains unclear. Arabic provides a challenging testbed due to its rich morphology and orthographic ambiguity, which create strong morphology-syntax interactions. We present a unified evaluation of LLMs on Arabic morphosyntactic tagging and dependency parsing, covering pre-tokenized, raw-text, and cascaded settings. We compare zero-shot prompting with retrieval-based in-context learning. Relevant demonstrations substantially improve performance. The strongest LLMs approach supervised tagging and parsing systems; however, they require substantial annotated data for demonstration retrieval and considerable computational resources. We make all code and data used in this paper publicly available.

cs.CL↗

arXiv2025

A highly sensitive, self-adhesive, biocompatible DLP 3D printed organohydrogel for flexible sensors and wearable devices

With the increasing demand for personalized health monitoring, wearable sensors have gained attention in medical diagnostics and physiological tracking. Hydrogels, known for their mechanical properties and similarity to biological tissues, are ideal for flexible sensing. However, conventional hydrogels face challenges in stability, biocompatibility, adhesion, and long-term comfort, especially in dynamic conditions.This study presents a highly sensitive, self-adhesive, and biocompatible organohydrogel fabricated via DLP 3D printing. By integrating an entanglement-dominated crosslinking mechanism with chemical and physical crosslinking, the hydrogel achieves high elasticity, mechanical strength, and durability. Methacrylic anhydride-grafted \k{appa}-carrageenan serves as the primary network, with optimized grafting rates enhancing tensile properties and strain modulation. The copolymer network of MA-kappa-CA and ACMO benefits from steric hindrance effects, improving swelling integrity and long-term stability.Experimental results confirm sustained adhesion and structural integrity under prolonged skin exposure, making it suitable for extended wear. The hydrogel exhibits excellent tensile resilience, flexibility, and strain-sensing capabilities. In vitro studies validate its biocompatibility, supporting its biomedical potential. Furthermore, its integration into wearable smart devices demonstrates promise for cervical spine monitoring and sports rehabilitation. A CNN-based system enables real-time, multi-channel analysis of cervical motion, proving its viability as a high-sensitivity flexible sensor for health monitoring and injury prevention.The proposed DLP 3D-printed hydrogel offers significant applications in flexible electronics, wearable sensors, and biomedical technologies, paving the way for next-generation health-monitoring systems.

physics.app-ph↗

arXiv2018

Text-Independent Speaker Verification Based on Deep Neural Networks and Segmental Dynamic Time Warping

In this paper we present a new method for text-independent speaker verification that combines segmental dynamic time warping (SDTW) and the d-vector approach. The d-vectors, generated from a feed forward deep neural network trained to distinguish between speakers, are used as features to perform alignment and hence calculate the overall distance between the enrolment and test utterances.We present results on the NIST 2008 data set for speaker verification where the proposed method outperforms the conventional i-vector baseline with PLDA scores and outperforms d-vector approach with local distances based on cosine and PLDA scores. Also score combination with the i-vector/PLDA baseline leads to significant gains over both methods.

cs.SD↗