Publications: OpenCaptchaWorld: A Comprehensive Web-based Platform for Testing and Benchmarking Multimodal LLM Agents (NeurIPS 2025); APL: Anchor-Based Prompt Learning for One-Stage Weakly Supervised Referring Expression Comprehension (ECCV 2024).
Research Experience
Currently a PhD student in Machine Learning at MBZUAI, closely working with Xiaofu Chen. Research focuses on developing unified understanding/generation and physically aware multimodal foundation models, making them efficient and deployable on edge devices.
Education
PhD: MBZUAI, Advisors: Prof. Zhiqiang Shen and Prof. Mohsen Guizani; Bachelor's Degree: Technical University of Denmark, Advisor: Prof. Dim P. Papadopoulos.
Background
Research Interests: Multimodal Foundation Model, Efficient Machine Learning, Physics Grounded Foundation Model. Professional Fields: Machine Learning, Deep Generative Modeling.
Miscellany
Personal Interests: Studied Biomedicine at the University of Queensland and managed a multi-brand boutique, studied pure mathematics and physics at the University of Edinburgh, which sparked a passion for science and technology.