arXiv subjects

Ahmed Jaber

Publications and source records attributed to Ahmed Jaber.

13 recordsLinked to original sources

arXiv2026

Microscopic theory of a radiation-balanced solar laser

We develop a microscopic open-quantum-system theory for a radiation-balanced solar laser (RBSL) based on ytterbium-doped yttrium aluminum garnet (Yb:YAG), in which optical gain, thermal redistribution among sublevels of the electronic ground and excited manifolds, and lattice-temperature dynamics are treated within a unified framework. Starting from a Lindblad master equation for a multilevel gain medium coupled to a cavity mode, we include incoherent solar pumping, spontaneous emission, cavity loss, and phonon-assisted intra-manifold relaxation obeying detailed balance. In the regime of fast thermalization within each electronic manifold, a compact temperature-dependent two-level model is derived, in which the gain, inversion, and lasing threshold are controlled by Boltzmann occupation factors and partition functions of the electronic sublevels. This microscopic reduction is then coupled self-consistently to a thermal balance equation accounting for anti-Stokes fluorescence cooling, quantum-defect heating, parasitic absorption, and heat exchange with the environment. The theory predicts several operating regimes, including pure cooling, lasing with net cooling, and lasing with net heating, as well as dynamical effects such as delayed lasing onset induced by self-cooling into threshold. In contrast to earlier radiation-balanced laser (RBL) models based mainly on macroscopic rate equations and thermodynamic balance arguments, the present approach provides a microscopic description of the feedback between quantum optical dynamics and temperature redistribution. It therefore offers a physically transparent framework for analyzing RBSLs and for identifying design strategies that exploit level structure, thermalization, and photonic-environment engineering to stabilize laser operation while minimizing internal heat load.

physics.optics↗

arXiv2026

AutoClimDS: Climate Data Science Agentic AI -- A Knowledge Graph is All You Need

Climate data science remains constrained by fragmented data sources, heterogeneous formats, and steep technical expertise requirements. These barriers slow discovery, limit participation, and undermine reproducibility. We present AutoClimDS, a Minimum Viable Product (MVP) Agentic AI system that addresses these challenges by integrating a curated climate knowledge graph (KG) with a set of Agentic AI workflows designed for cloud-native scientific analysis. The KG unifies datasets, metadata, tools, and workflows into a machine-interpretable structure, while AI agents, powered by generative models, enable natural-language query interpretation, automated data discovery, programmatic data acquisition, and end-to-end climate analysis. A key result is that AutoClimDS can reproduce published scientific figures and analyses from natural-language instructions alone, completing the entire workflow from dataset selection to preprocessing to modeling. When given the same tasks, state-of-the-art general-purpose LLMs (e.g., ChatGPT GPT-5.1) cannot independently identify authoritative datasets or construct valid retrieval workflows using standard web access. This highlights the necessity of structured scientific memory for agentic scientific reasoning. By encoding procedural workflow knowledge into a KG and integrating it with existing technologies (cloud APIs, LLMs, sandboxed execution), AutoClimDS demonstrates that the KG serves as the essential enabling component, the irreplaceable structural foundation, for autonomous climate data science. This approach provides a pathway toward democratizing climate research through human-AI collaboration.

cs.AI↗

arXiv2025

Enhancing Pavement Crack Classification with Bidirectional Cascaded Neural Networks

Pavement distress, such as cracks and potholes, is a significant issue affecting road safety and maintenance. In this study, we present the implementation and evaluation of Bidirectional Cascaded Neural Networks (BCNNs) for the classification of pavement crack images following image augmentation. We classified pavement cracks into three main categories: linear cracks, potholes, and fatigue cracks on an enhanced dataset utilizing U-Net 50 for image augmentation. The augmented dataset comprised 599 images. Our proposed BCNN model was designed to leverage both forward and backward information flows, with detection accuracy enhanced by its cascaded structure wherein each layer progressively refines the output of the preceding one. Our model achieved an overall accuracy of 87%, with precision, recall, and F1-score measures indicating high effectiveness across the categories. For fatigue cracks, the model recorded a precision of 0.87, recall of 0.83, and F1-score of 0.85 on 205 images. Linear cracks were detected with a precision of 0.81, recall of 0.89, and F1-score of 0.85 on 205 images, and potholes with a precision of 0.96, recall of 0.90, and F1-score of 0.93 on 189 images. The macro and weighted average of precision, recall, and F1-score were identical at 0.88, confirming the BCNN's excellent performance in classifying complex pavement crack patterns. This research demonstrates the potential of BCNNs to significantly enhance the accuracy and reliability of pavement distress classification, resulting in more effective and efficient pavement maintenance and management systems.

cs.CV↗

arXiv2025

Zero-Shot Scene Understanding with Multimodal Large Language Models for Automated Vehicles

Scene understanding is critical for various downstream tasks in autonomous driving, including facilitating driver-agent communication and enhancing human-centered explainability of autonomous vehicle (AV) decisions. This paper evaluates the capability of four multimodal large language models (MLLMs), including relatively small models, to understand scenes in a zero-shot, in-context learning setting. Additionally, we explore whether combining these models using an ensemble approach with majority voting can enhance scene understanding performance. Our experiments demonstrate that GPT-4o, the largest model, outperforms the others in scene understanding. However, the performance gap between GPT-4o and the smaller models is relatively modest, suggesting that advanced techniques such as improved in-context learning, retrieval-augmented generation (RAG), or fine-tuning could further optimize the smaller models' performance. We also observe mixed results with the ensemble approach: while some scene attributes show improvement in performance metrics such as F1-score, others experience a decline. These findings highlight the need for more sophisticated ensemble techniques to achieve consistent gains across all scene attributes. This study underscores the potential of leveraging MLLMs for scene understanding and provides insights into optimizing their performance for autonomous driving applications.

cs.CV↗

arXiv2025

Vision-Language Models for Autonomous Driving: CLIP-Based Dynamic Scene Understanding

Scene understanding is essential for enhancing driver safety, generating human-centric explanations for Automated Vehicle (AV) decisions, and leveraging Artificial Intelligence (AI) for retrospective driving video analysis. This study developed a dynamic scene retrieval system using Contrastive Language-Image Pretraining (CLIP) models, which can be optimized for real-time deployment on edge devices. The proposed system outperforms state-of-the-art in-context learning methods, including the zero-shot capabilities of GPT-4o, particularly in complex scenarios. By conducting frame-level analysis on the Honda Scenes Dataset, which contains a collection of about 80 hours of annotated driving videos capturing diverse real-world road and weather conditions, our study highlights the robustness of CLIP models in learning visual concepts from natural language supervision. Results also showed that fine-tuning the CLIP models, such as ViT-L/14 and ViT-B/32, significantly improved scene classification, achieving a top F1 score of 91.1%. These results demonstrate the ability of the system to deliver rapid and precise scene recognition, which can be used to meet the critical requirements of Advanced Driver Assistance Systems (ADAS). This study shows the potential of CLIP models to provide scalable and efficient frameworks for dynamic scene understanding and classification. Furthermore, this work lays the groundwork for advanced autonomous vehicle technologies by fostering a deeper understanding of driver behavior, road conditions, and safety-critical scenarios, marking a significant step toward smarter, safer, and more context-aware autonomous driving systems.

cs.CV↗

arXiv2024

Advancing Object Detection in Transportation with Multimodal Large Language Models (MLLMs): A Comprehensive Review and Empirical Testing

This study aims to comprehensively review and empirically evaluate the application of multimodal large language models (MLLMs) and Large Vision Models (VLMs) in object detection for transportation systems. In the first fold, we provide a background about the potential benefits of MLLMs in transportation applications and conduct a comprehensive review of current MLLM technologies in previous studies. We highlight their effectiveness and limitations in object detection within various transportation scenarios. The second fold involves providing an overview of the taxonomy of end-to-end object detection in transportation applications and future directions. Building on this, we proposed empirical analysis for testing MLLMs on three real-world transportation problems that include object detection tasks namely, road safety attributes extraction, safety-critical event detection, and visual reasoning of thermal images. Our findings provide a detailed assessment of MLLM performance, uncovering both strengths and areas for improvement. Finally, we discuss practical limitations and challenges of MLLMs in enhancing object detection in transportation, thereby offering a roadmap for future research and development in this critical area.

cs.CV↗

arXiv2024

Visual Reasoning and Multi-Agent Approach in Multimodal Large Language Models (MLLMs): Solving TSP and mTSP Combinatorial Challenges

Multimodal Large Language Models (MLLMs) harness comprehensive knowledge spanning text, images, and audio to adeptly tackle complex problems, including zero-shot in-context learning scenarios. This study explores the ability of MLLMs in visually solving the Traveling Salesman Problem (TSP) and Multiple Traveling Salesman Problem (mTSP) using images that portray point distributions on a two-dimensional plane. We introduce a novel approach employing multiple specialized agents within the MLLM framework, each dedicated to optimizing solutions for these combinatorial challenges. Our experimental investigation includes rigorous evaluations across zero-shot settings and introduces innovative multi-agent zero-shot in-context scenarios. The results demonstrated that both multi-agent models. Multi-Agent 1, which includes the Initializer, Critic, and Scorer agents, and Multi-Agent 2, which comprises only the Initializer and Critic agents; significantly improved solution quality for TSP and mTSP problems. Multi-Agent 1 excelled in environments requiring detailed route refinement and evaluation, providing a robust framework for sophisticated optimizations. In contrast, Multi-Agent 2, focusing on iterative refinements by the Initializer and Critic, proved effective for rapid decision-making scenarios. These experiments yield promising outcomes, showcasing the robust visual reasoning capabilities of MLLMs in addressing diverse combinatorial problems. The findings underscore the potential of MLLMs as powerful tools in computational optimization, offering insights that could inspire further advancements in this promising field. Project link: https://github.com/ahmed-abdulhuy/Solving-TSP-and-mTSP-Combinatorial-Challenges-using-Visual-Reasoning-and-Multi-Agent-Approach-MLLMs-.git

cs.AI↗

arXiv2024

Object Detection using Oriented Window Learning Vi-sion Transformer: Roadway Assets Recognition

Object detection is a critical component of transportation systems, particularly for applications such as autonomous driving, traffic monitoring, and infrastructure maintenance. Traditional object detection methods often struggle with limited data and variability in object appearance. The Oriented Window Learning Vision Transformer (OWL-ViT) offers a novel approach by adapting window orientations to the geometry and existence of objects, making it highly suitable for detecting diverse roadway assets. This study leverages OWL-ViT within a one-shot learning framework to recognize transportation infrastructure components, such as traffic signs, poles, pavement, and cracks. This study presents a novel method for roadway asset detection using OWL-ViT. We conducted a series of experiments to evaluate the performance of the model in terms of detection consistency, semantic flexibility, visual context adaptability, resolution robustness, and impact of non-max suppression. The results demonstrate the high efficiency and reliability of the OWL-ViT across various scenarios, underscoring its potential to enhance the safety and efficiency of intelligent transportation systems.

cs.CV↗

arXiv2024

Eyeballing Combinatorial Problems: A Case Study of Using Multimodal Large Language Models to Solve Traveling Salesman Problems

Multimodal Large Language Models (MLLMs) have demonstrated proficiency in processing di-verse modalities, including text, images, and audio. These models leverage extensive pre-existing knowledge, enabling them to address complex problems with minimal to no specific training examples, as evidenced in few-shot and zero-shot in-context learning scenarios. This paper investigates the use of MLLMs' visual capabilities to 'eyeball' solutions for the Traveling Salesman Problem (TSP) by analyzing images of point distributions on a two-dimensional plane. Our experiments aimed to validate the hypothesis that MLLMs can effectively 'eyeball' viable TSP routes. The results from zero-shot, few-shot, self-ensemble, and self-refine zero-shot evaluations show promising outcomes. We anticipate that these findings will inspire further exploration into MLLMs' visual reasoning abilities to tackle other combinatorial problems.

cs.AI↗

arXiv2024

Hybrid THz architectures for molecular polaritonics

Physical and chemical properties of materials can be modified by a resonant optical mode. Such recent demonstrations have mostly relied on a planar cavity geometry, others have relied on a plasmonic resonator. However, the combination of these two device architectures have remained largely unexplored, especially in the context of maximizing light-matter interactions. Here, we investigate several schemes of electromagnetic field confinement aimed at facilitating the collective coupling of a localized photonic mode to molecular vibrations in the terahertz region. The key aspects are the use of metasurface plasmonic structures combined with standard Fabry-Perot configurations and the deposition of a thin layer of glucose, via a spray coating technique, within a tightly focused electromagnetic mode volume. More importantly, we demonstrate enhanced vacuum Rabi splittings reaching up to 200 GHz when combining plasmonic resonances, photonic cavity modes and low-energy molecular resonances. Furthermore, we demonstrate how a cavity mode can be utilized to enhance the zero-point electric field amplitude of a plasmonic resonator. Our study provides key insight into the design of polaritonic platforms with organic molecules to harvest the unique properties of hybrid light-matter states.

physics.optics↗

arXiv2022

Metamaterial-based octave-wide terahertz bandpass filters

We present octave-wide bandpass filters in the terahertz (THz) region based on bilayer-metamaterial (BLMM) structures. The passband region has a super-Gaussian shape with a maximum transmittance approaching 70% and a typical stopband rejection of 20 dB. The design is based on a metasurface consisting of a metallic square-hole array deposited on a transparent polymer, which is stacked on top of an identical metasurface with a sub-wavelength separation. The superimposed metasurface structures were designed using finite-difference time-domain (FDTD) simulations and fabricated using a photolithography process. Experimental characterization of these structures between 0.3 to 5.8 THz is performed with a time-domain THz spectroscopy system. Good agreement between experiment and simulation results is observed. We also demonstrate that two superimposed BLMM (2BLMM) devices increase the steepness of the roll-offs to more than 85 dB/octave and enable a superior stopband rejection approaching 40 dB while the maximum transmittance remains above 64%. This work paves the way toward new THz applications, including the detection of THz pulses centered at specific frequencies, and an enhanced time-resolved detection sensitivity towards molecular vibrations that are noise dominated by a strong, off-resonant, driving field.

physics.optics↗

arXiv2021

An Application of BnB-NSGAII: Initializing NSGAII to Solve 3 Stage Reducer Problem

The 3 stage reducer problem is a point of interest for many researchers. In this paper, this problem is reformulated to a bi-objective problem with additional constraints to meet the ISO mechanical standards. Those additional constraints increase the complexity of the problem, such that, NSGAII performance is not sufficient. To overcome this, we propose to use BnB-NSGAII method - a hybrid multi-criteria branch and bound with NSGAII - to initialize NSGAII before solving the problem, seeking for a better initial population. A new feature is also proposed to enhance BnB-NSGAII method, called the legacy feature. The legacy feature permits the inheritance of the elite individuals between - branch and bound - parent and children nodes. NSGAII and BnB-NSGAII with and without the legacy feature are tested on the 3 stage reducer problem. Results demonstrate the competitive performance of BnB-NSGAII with the legacy feature.

math.OC↗

arXiv2020

A Branch and Bound Based on NSGAII Algorithm for Multi-Objective Mixed Integer Non Linear Optimization Problems

Multi-Objective Mixed-Integer Non-Linear Programming problems (MO-MINLPs) appear in several real-world applications, especially in the mechanical engineering field. To determine a good approximated Pareto front for this type of problems, we propose a general hybrid approach based on a Multi-Criteria Branch-and-Bound (MCBB) and Non-dominated Sorting Genetic Algorithm 2 (NSGAII). We present a computational experiment based on a statistical assessment to compare the performance of the proposed algorithm (BnB-NSGAII) with NSGAII using well-known metrics from literature. We propose a new metric, Investment Ratio (IR), that relate the quality of the solution to the consumed effort. We consider five real-world mechanical engineering problems and two mathematical ones to be used as test problems in this experiment. Experimental results indicate that BnB-NSGAII could be a competitive alternative for solving MO-MINLPs.

math.OC↗