Elling Svee

About

I am a PhD student at NTNU working on applied mathematics and statistics. My research interests include spatio-temporal modeling, HPC and machine learning.

Research

Various Projects

CO2IPSimulator: Graph-based modeling of CO2 migration.
sparsejax: Sparse operations in JAX.
JPI: Distributed computing with JAX using MPI.
JutulGPT: An AI assistant for the JutulDarcy.jl reservoir simulator.

Publications

Bayesian inversion of multilayer CO2 migration from seismic plume observations using a graph-based finite-rate invasion-percolation model
Elling Svee, Jo Eidsvik, Kasper Hunnestad and Philip Ringrose
Preprint, 2026
arXiv

Agentic Scientific Simulation: Execution-Grounded Model Construction and Reconstruction
Knut-Andreas Lie, Olav Møyner, Elling Svee and Jakob Torben
Preprint, 2026
arXiv

Non-stationary Spatial Modeling Using Fractional SPDEs
Elling Svee and Geir-Arne Fuglstad
Published, 2026
Spatial Statistics, arXiv

A Framework for Spatial Modeling Using Fractional and Non-Stationary SPDEs
Elling Svee
Masters Thesis, 2025
Nasjonalt vitenarkiv

Teaching

Autumn 2026: Assistant in TMA4265 Stochastic Modeling. See course materials.
Spring 2026: Assistant in TMA4320 Introduction to Scientific Computation.

Other

Blog

Posterior Sampling with Conditional Diffusion Models: Using score-based diffusion models and training-free guidance for Bayesian inversion.

Posters

Geilo Winter School January 2026: CO2BatchFill.jl: Efficient modeling of multi-layer CO2 migration (PDF)

Posts

SINTEF-blog: Talking to Your Simulator: What We Learned Building JutulGPT

Contact

Email: elling.svee@ntnu.no
Address:
1056, Sentralbygg 2
NTNU Gløshaugen
Trondheim, Norway