Point cloud-based diffusion models for the Electron-Ion Collider

TL;DR

This paper introduces a novel point cloud-based diffusion model for generating complete collider events at the Electron-Ion Collider, improving fidelity and efficiency over previous image-based approaches and leveraging foundation models for high-energy physics simulations.

Contribution

It presents a new point cloud diffusion architecture for full event generation, incorporating event-wide constraints and adapting foundation models for collider data.

Findings

Successfully generates entire collider events with conservation laws

Addresses data sparsity issues with point cloud and transformer techniques

Demonstrates potential for foundation models in high-energy physics simulations

Abstract

At high-energy collider experiments, generative models can be used for a wide range of tasks, including fast detector simulations, unfolding, searches of physics beyond the Standard Model, and inference tasks. In particular, it has been demonstrated that score-based diffusion models can generate high-fidelity and accurate samples of jets or collider events. This work expands on previous generative models in three distinct ways. First, our model is trained to generate entire collider events, including all particle species with complete kinematic information. We quantify how well the model learns event-wide constraints such as the conservation of momentum and discrete quantum numbers. We focus on the events at the future Electron-Ion Collider, but we expect that our results can be extended to proton-proton and heavy-ion collisions. Second, previous generative models often relied on image-based techniques. The sparsity of the data can negatively affect the fidelity and sampling time of the model. We address these issues using point clouds and a novel architecture combining edge creation with transformer modules called Point Edge Transformers. Third, we adapt the foundation model OmniLearn, to generate full collider events. This approach may indicate a transition toward adapting and fine-tuning foundation models for downstream tasks instead of training new models from scratch.