Set-Conditional Set Generation for Particle Physics

TL;DR

This paper introduces a novel graph neural network-based generative model with slot-attention for efficient simulation of set-valued particle physics data, significantly outperforming existing methods.

Contribution

The paper presents a new generative model combining graph neural networks and slot-attention, tailored for conditional set generation in particle physics simulations.

Findings

Outperforms existing baseline models in simulation accuracy

Reduces computational time for particle physics data generation

Demonstrates effectiveness on Large Hadron Collider data

Abstract

The simulation of particle physics data is a fundamental but computationally intensive ingredient for physics analysis at the Large Hadron Collider, where observational set-valued data is generated conditional on a set of incoming particles. To accelerate this task, we present a novel generative model based on a graph neural network and slot-attention components, which exceeds the performance of pre-existing baselines.