Extrapolating Jet Radiation with Autoregressive Transformers

TL;DR

This paper introduces autoregressive transformers to generate variable-particle jet events in high-energy physics, enabling extrapolation to higher multiplicities and improving modeling of QCD jet radiation.

Contribution

It demonstrates how transformers can learn a factorized likelihood for jet radiation and extrapolate to higher multiplicities, a novel approach in this domain.

Findings

Transformers successfully model jet radiation likelihoods.

Effective extrapolation to higher jet multiplicities.

Training modifications improve likelihood learning.

Abstract

Generative networks are an exciting tool for fast LHC event fixed number of particles. Autoregressive transformers allow us to generate events containing variable numbers of particles, very much in line with the physics of QCD jet radiation, and offer the possibility to generalize to higher multiplicities. We show how transformers can learn a factorized likelihood for jet radiation and extrapolate in terms of the number of generated jets. For this extrapolation, bootstrapping training data and training with modifications of the likelihood loss can be used.