This dataset contains a detailed simulation output that allows the construction and study of different data representations for electromagnetic and hadronic showers in calorimeters. It is published so that optimal data representations can be studied, with the ultimate goal of constructing a general tool that takes detailed simulation output and translates it into an optimal representation that can serve as the input to surrogate simulators based on generative models.
@article{arxiv.2509.22340,
title = {step2point dataset: Detailed shower simulation for data representation studies},
author = {Anna Zaborowska and Peter McKeown},
journal= {arXiv preprint arXiv:2509.22340},
year = {2025}
}