# L2LFlows: Generating High-Fidelity 3D Calorimeter Images

## TL;DR

L2LFlows employs layered normalizing flows conditioned on previous layers to generate high-fidelity 3D calorimeter images, significantly improving over existing generative models in simulating photon showers.

## Contribution

The paper introduces Layer-to-Layer-Flows, a novel high-dimensional normalizing flow architecture conditioned on multiple layers for improved calorimeter image generation.

## Key findings

- L2LFlows outperforms BIB-AE in image fidelity.
- The model effectively captures layer-to-layer correlations.
- High-dimensional normalizing flows are feasible for detailed detector simulations.

## Abstract

We explore the use of normalizing flows to emulate Monte Carlo detector simulations of photon showers in a high-granularity electromagnetic calorimeter prototype for the International Large Detector (ILD). Our proposed method -- which we refer to as "Layer-to-Layer-Flows" (L$2$LFlows) -- is an evolution of the CaloFlow architecture adapted to a higher-dimensional setting (30 layers of $10\times 10$ voxels each). The main innovation of L$2$LFlows consists of introducing $30$ separate normalizing flows, one for each layer of the calorimeter, where each flow is conditioned on the previous five layers in order to learn the layer-to-layer correlations. We compare our results to the BIB-AE, a state-of-the-art generative network trained on the same dataset and find our model has a significantly improved fidelity.