Katarina Petrovic
Katarina Petrovic
Biography
My research focuses on flow-based generative models, particularly flow matching and diffusion, to capture the evolution of complex dynamical systems across space and time. I am particularly interested in AI4Science applications such as modelling cell proliferation and development of microenvironments.
I previously completed a research internship at Network Science Institute in Boston in Barabasi Lab, where I focused on leveraging LLMs to identify potential drug repurposing candidates and learn underlying disease mechanisms in gene networks.
I hold MEng in Engineering Science from University of Oxford, where I was a recipient of Simon and June Li full-ride Oxford scholarship awarded on the basis of outstanding academic merit. I completed my Master’s thesis at Oxford Man Institute of Quantitative Finance under supervision of Prof Xiaowen Dong and Dr Jan-Peter Calliess, where I was developing multi-period optimization algorithms for trading in the FX markets.
Selected Publications
-
Curly Flow Matching for Learning Non−gradient Field Dynamics
Katarina Petrović‚ Lazar Atanackovic‚ Viggo Moro‚ Kacper Kapuśniak‚ İsmail İlkan Ceylan‚ Michael Bronstein‚ Michael Bronstein‚ Avishek Joey Bose and Alexander Tong
In Proceedings of the 39th Annual Conference on Neural Information Processing Systems (NeurIPS). 2025.
Details about Curly Flow Matching for Learning Non−gradient Field Dynamics | BibTeX data for Curly Flow Matching for Learning Non−gradient Field Dynamics