Diffusion-Based Point-Cloud Generation of Heavy-Ion Events

TL;DR

This paper introduces a fast, high-fidelity generative model for simulating heavy-ion collision events using a diffusion process and transformer architecture, enabling realistic high-multiplicity event generation.

Contribution

The paper presents a novel diffusion-based generative model with a two-stage training strategy for realistic heavy-ion event simulation, improving efficiency and fidelity.

Findings

Model accurately reproduces event- and particle-level observables.

Generates realistic high-multiplicity heavy-ion events.

Achieves promising results in jet finding and classifier-based fidelity tests.

Abstract

Heavy-ion collisions produce final states with thousands to tens of thousands of particles, making their simulation among the most computationally intensive tasks in high-energy nuclear physics. We present a fast, high-fidelity generative model for heavy-ion events based on a score-driven diffusion process and the Point-Edge Transformer architecture within the OmniLearn framework. A two-stage training strategy is performed: Stage-1 training on lower-multiplicity O-O collisions allowing the model to learn a stable event and particles representation, followed by fine-tuning on challenging high-multiplicity Pb-Pb collisions. We benchmark the generator with a broad set of closure checks, including agreement of event- and particle-level observables in one and two dimensions, flow consistency reconstructed from the generated particles, end-to-end jet finding with FastJet including key jet and substructure observables, and a classifier-based application to quantify the sample fidelity. The results are promising, showing that a compact generative model can produce realistic, high-multiplicity heavy-ion events, at a level that makes local-scale generation for heavy-ion collisions at high energies a practical goal.