Research
My work connects language models, scientific reasoning, and physical systems. It spans both current research in AI for science and an earlier body of work in computational fashion.
Imperial College London
I study how language models behave as surrogate models in low-data optimization. The work shows that prompt design and query protocol are core components of surrogate specification, shaping uncertainty, belief updating, and downstream scientific decisions.
Imperial College London
I develop LLM agents for battery fast-charging optimization. The agents expand protocol search beyond fixed human parameterizations, incorporate natural-language safety constraints, and operate in expensive, slow-feedback scientific settings.
Imperial College London
I investigate how LLMs encode structured scientific knowledge. This work identifies a 3D spiral structure in hidden states that aligns with the conceptual organization of the periodic table.
Imperial College London
I build automated systems for extracting, structuring, and operationalizing scientific knowledge from literature and imaging workflows, including MicroGPT for 3D microstructure analysis.
Donghua University
I developed computer-vision methods for mapping flexible garment surfaces and translating 3D shape information into structured representations for apparel development.
Donghua University
I explored how individual body features and user-specific design needs can be translated into intelligent made-to-measure and apparel development workflows.
Donghua University
I investigated computational approaches to garment pattern-making that reduce waste while preserving fit, function, manufacturability, and design intent.