SciPost logo

SciPost Submission Page

PC-JeDi: Diffusion for Particle Cloud Generation in High Energy Physics

by Matthew Leigh, Debajyoti Sengupta, Guillaume Quétant, John Andrew Raine, Knut Zoch, Tobias Golling

This Submission thread is now published as

Submission summary

Authors (as registered SciPost users): John Raine · Debajyoti Sengupta
Submission information
Preprint Link: scipost_202309_00013v2  (pdf)
Date accepted: 2023-12-27
Date submitted: 2023-12-12 14:14
Submitted by: Raine, John
Submitted to: SciPost Physics
Ontological classification
Academic field: Physics
  • High-Energy Physics - Experiment
  • High-Energy Physics - Phenomenology
Approach: Phenomenological


In this paper, we present a new method to efficiently generate jets in High Energy Physics called PC-JeDi. This method utilises score-based diffusion models in conjunction with transformers which are well suited to the task of generating jets as particle clouds due to their permutation equivariance. PC-JeDi achieves competitive performance with current state-of-the-art methods across several metrics that evaluate the quality of the generated jets. Although slower than other models, due to the large number of forward passes required by diffusion models, it is still substantially faster than traditional detailed simulation. Furthermore, PC-JeDi uses conditional generation to produce jets with a desired mass and transverse momentum for two different particles, top quarks and gluons.

Author comments upon resubmission

We thank the reviewers for the follow up feedback.
We have now calculated uncertainties for the remaining metrics and included them in the tables of results.
Other minor changes have been made, but not which change the results or interpretation thereof.

Best regards

List of changes

Include uncertainties on additional metrics as requested.
Fix of the calculation in the D2 observable
Renaming of the simulated dataset in plots to reflect the simulation chain (previously incorrectly labelled as delphes)
Add "upper limit" of metrics for performance, calculated from reference monte carlo simulation
Updated precision in calculation of metrics (using larger number of samples)

Published as SciPost Phys. 16, 018 (2024)

Login to report or comment