Is Tokenization Needed for Masked Particle Modelling?

TL;DR

This paper advances masked particle modeling by introducing new reconstruction methods that outperform tokenized approaches, enhancing the development of foundation models for high-energy physics without data discretization.

Contribution

It presents novel reconstruction techniques using conditional generative models that improve masked particle modeling performance in high-energy physics applications.

Findings

New methods outperform tokenized MPM on jet physics tasks

Enhanced decoder and implementation efficiencies improve model performance

Applicable to various downstream tasks like classification and vertex finding

Abstract

In this work, we significantly enhance masked particle modeling (MPM), a self-supervised learning scheme for constructing highly expressive representations of unordered sets relevant to developing foundation models for high-energy physics. In MPM, a model is trained to recover the missing elements of a set, a learning objective that requires no labels and can be applied directly to experimental data. We achieve significant performance improvements over previous work on MPM by addressing inefficiencies in the implementation and incorporating a more powerful decoder. We compare several pre-training tasks and introduce new reconstruction methods that utilize conditional generative models without data tokenization or discretization. We show that these new methods outperform the tokenized learning objective from the original MPM on a new test bed for foundation models for jets, which includes using a wide variety of downstream tasks relevant to jet physics, such as classification, secondary vertex finding, and track identification.