Machine Learning, LLM Safety, Adversarial Machine Learning, Sequential Change-Point Detection.
Mohammed Alshaalan is a PhD candidate at the Department of Electronic and Electrical Engineering at University College London (UCL), working on the safety and security of large language models, with a current focus on detecting adversarial attacks during inference using sequential statistical methods. His PhD is sponsored by Saudi Aramco's EXPEC Advance Research Center, where he previously worked for seven years on applied machine learning and information security. He earned his Master's in Applied Data Analytics from Boston University and a Bachelor's in Software Engineering from King Fahd University of Petroleum and Minerals (KFUPM).