Track finding and particle identification performance with particle physics foundation models
Authors: Shuhang Li, Haiwang Yu, Gabor Galgoczi, Yeonju Go, Yi Huang, David Park, Michael Peters, Christopher Pinkenburg, Christof Roland, Jin Huang, Yihui Ren, Joseph D. Osborn
Organizations: Columbia University, New York, NY 10027, USA · Brookhaven National Laboratory, Upton, NY 11973, USA · State University of New York, Stony Brook, New York, NY 11794, USA · Massachusetts Institute of Technology, Cambridge, MA 02139, USA
Modern high energy nuclear and particle physics experiments produce large amounts of sparse data that must be reduced significantly in size before performing physics analysis. Recently, there has been an effort to evaluate Foundation Models as a potential tool for performing data analysis. These models have been shown to learn features in high dimensional datasets based on the multi-modal data they are trained on. To show their effectiveness at performing complex pattern recognition tasks, such as reconstructing charged particles in a magnetic field, we evaluated the performance of a Foundation Model based reconstruction on several downstream tasks with a dataset composed of spacepoints in a Time Projection Chamber. We show that the performance of the Foundation Model based data analysis significantly improves upon that of traditional methods, and furthermore can be adapted to learn additional tasks that are not necessarily related to the initial training objective.
Figures & tables
Figure 1: The track finding efficiency for the FM and the Cellular Automaton based seeding is shown as a function of pT . The FM based seeding algorithm surpasses the CA seeder performance at low pT by a significant margin.
Figure 2: The track finding purity for the FM and the Cellular Automaton based seeding is shown as a function of pT .
Figure 3: One-vs-rest PID ROC curves for pions (left), kaons (middle), and protons (right) in the momentum slice p∈[0.8,1.2) GeV/ c , comparing the FM4NPP model (blue) with the traditional GPR dE/dx baseline (red). The dashed diagonal denotes random performance, and the area under each curve (AUC) is quoted in the legend. Both methods are evaluated on the same reconstructed tracks matched to primary particles. Tracks are charge-folded and the classification is restricted to the π/K/p hypotheses for both methods.
Department of Physics, Stanford University, Stanford, CA 94305, USA · Nagoya University, Kobayashi-Maskawa Institute, Aichi 464-8602, Japan · Fundamental Physics Directorate, SLAC National Accelerator Laboratory, Menlo Park, CA 94025, USA +1
University of California San Diego, La Jolla, California 92093, USA · National Institute of Chemical Physics and Biophysics, Tallinn 12618, Estonia · SLAC National Accelerator Laboratory, Menlo Park, California 94025, USA