Md Kamrul Islam

About Me

I am a Lecturer in the Department of Computer Science at American International University-Bangladesh (AIUB), with research interests spanning Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), Geometric Deep Learning, and AI in Healthcare.

I have completed the Erasmus Mundus Joint Master Degree in Big Data Management and Analytics (BDMA), studying across CentraleSupélec, Universitat Politècnica de Catalunya (UPC), and Université Libre de Bruxelles (ULB). For my master’s thesis, supervised by Dr. Tiphaine Henry, I worked on Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) for security requirement extraction from Business Process Model and Notation (BPMN 2.0), leading to a publication at the BPM Forum 2026. Previously, I earned a Bachelor of Engineering in Software Engineering from Sichuan University.

My current research interests lie in Multimodal Large Language Models (MLLMs), AI in Healthcare, Frugal LLMs, and Retrival-augmented Generation (RAG). When I’m not working, I enjoy practicing new languages or playing tennis.

Always happy to connect, exchange ideas, and explore potential collaborations.

mdkamrul.islam@aiub.edu / mdkamrul.islam@hotmail.com

Research Highlights

My academic and research experiences reflect this exploration:

• Master’s Thesis: A Hybrid LLM-Based Framework for Automated Security Annotation Generation in Business Process Models, accepted at the BPM Forum of the 24th International Conference on Business Process Management (BPM 2026). [Preprint]

• Graduate Research Assistant: Developed unsupervised deep clustering frameworks based on group-equivariant CNNs using large scale Chest X-ray Dataset to improve feature representation and cluster separability while reducing reliance on data augmentation. See project: Unsupervised chest X-ray clustering with group-equivariant convolutions.

• Capstone Project: Designed and trained CNN architectures for multi-class tumor classification from brain MRI scans, with a focus on preprocessing, evaluation metrics, and robustness. See project: Brain Tumor Detection and Classification Using Convolutional Neural Networks.

Expertise

• Core ML: Geometric Deep Learning, Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), Information Extraction

• Domains: AI in healthcare, security-aware business process modeling, Knowledge Graphs

Hobbies

Language Exchange · Reading · Tennis · Cycling · Traveling · Hiking