Geant4-FastSim - Fast inference of Diffusion models

Fast and accurate simulation of high-energy physics experiments is crucial for advancing our understanding of fundamental particles and forces in nature. However, the traditional method of obtaining such simulations using Monte Carlo methods is computationally expensive. Recently, machine learning techniques such as generative modeling have been proposed as an alternative approach to providing simulation results. In particular, diffusion models have emerged as highly accurate, but the standard formulation of the diffusion process suffers from slow inference speeds. This project aims to accelerate the inference of diffusion models by exploring recently proposed techniques in the field, such as distillation, continuous-time diffusion formulations, or more efficient architectures. The methods developed in this project will be integrated into the Geant4FasSim repository, enhancing its capabilities for fast and accurate simulation of high-energy physics experiments.