hep-exOct 11, 2026

Using Line Segment Tracking to enable ML-based Track Reconstruction in CMS

Authors: Aashay Arora on behalf of the CMS Collaboration

Organizations: University of California San Diego, La Jolla, CA, USA

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

High-pileup conditions in CMS during the HL-LHC era make charged-particle tracking increasingly challenging as detector occupancy and combinatorics grow. We present a hybrid approach that exploits Line Segment Tracking (LST) objects rather than individual hits to enable the first CMS ML-based track reconstruction algorithm. The LST objects, namely the triplets used in this study, are built according to geometry- and physics-driven criteria and carry richer local structure, geometrical cues, and physics quantities such as transverse momentum, which helps the model incorporate detector-level information from the start. We apply a Transformer encoder with geometrically masked sparse self-attention to embed these objects in a learned latent space, and use object condensation to assemble full tracks in a single inference step. This object-based representation reduces complexity in dense environments and makes the model less sensitive to ambiguities introduced by high pileup. Evaluated on CMS Phase-2 simulation of ttˉt\bar{t} events at ⟨μ⟩=200\langle μ\rangle = 200, the approach achieves reconstruction efficiency comparable to the LST quintuplet baseline while reducing the fake rate by roughly a factor of three and suppressing duplicates almost entirely, demonstrating the promise of advanced ML architectures for next-generation particle tracking.

Figures & tables

Figure 1: Construction of an LST T3 and its underlying objects in the Phase-2 Outer Tracker, shown in the rr – zz plane
Figure 2: Architecture of the Transformer encoder with geometric sparse attention.
Figure 3: The track-reconstruction pipeline
Figure 4: For ttˉt\bar{t} events at ⟨μ⟩=200\langle\mu\rangle=200 . Left: condensation likelihood β\beta for real and fake T3s, with the operating point β≥0.005\beta\geq 0.005 indicated. Centre: ROC curve for β\beta as a real/fake T3 classifier; the marker is the operating point, retaining 80.2%80.2\% of true triplets at 2.0%2.0\% fake-triplet efficiency. Right: number of unique OT layers NLN_{L} spanned by the reconstructed candidates.
Figure 5: Tracking efficiency (left), fake rate (centre) and duplicate rate (right) versus η\eta (top) and pTp_{\mathrm{T}} (bottom) for the Transformer ++ object-condensation candidates (blue) and the LST T5 baseline (red), under the matched NL≥5N_{L}\geq 5 requirement.
Figure 6: As Figure 5 , but with the relaxed NL≥4N_{L}\geq 4 requirement applied to the Transformer ++ object-condensation candidates; the LST T5 baseline is unchanged.

Explore similar work

CardsList

  1. HEPTv2: End-to-End Efficient Point Transformer for Charged Particle Reconstruction

    Jun 18, 2026Siqi Miao, Shitij Govil, Jack P. Rodgers +5Efficient Transformer InferenceTransformer

    School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USA. · Department of Physics and Astronomy, Purdue University, West Lafayette, IN, USA. · Department of Physics, University of California San Diego, La Jolla, CA, USA. +1

  2. Vision Transformers and Graph Neural Networks for Charged Particle Tracking in the ATLAS Muon Spectrometer

    Mar 26, 2026Jonathan RenuschHigh-Energy PhysicsGraph Neural Networks

    on behalf of the ATLAS Collaboration · CERN, Geneva, Switzerland

  3. PhyGHT: Physics-Guided HyperGraph Transformer for Signal Purification at the HL-LHC

    Feb 24, 2026Mohammed Rakib, Luke Vaughan, Shivang Patel +3Physics-Informed GNNsHigh-Energy Physics

    Department of Computer Science Oklahoma State University, USA · Department of Physics Oklahoma State University, USA