Full event interpretation with machine-learning-based particle-flow reconstruction in the CMS detector


Sosnov D., Oreshkin V., Murzin V., Kim V., Dudko L., Snigirev A., ...Daha Fazla

European Physical Journal C, cilt.86, sa.8, 2026 (SCI-Expanded, Scopus)

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 86 Sayı: 8
  • Basım Tarihi: 2026
  • Doi Numarası: 10.1140/epjc/s10052-026-15754-5
  • Dergi Adı: European Physical Journal C
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Aerospace Database, Chemical Abstracts Core, Compendex, INSPEC, zbMATH, Directory of Open Access Journals, Academic Search Ultimate (EBSCO), Engineering Source (EBSCO), Technology Collection (ProQuest)
  • Yozgat Bozok Üniversitesi Adresli: Evet

Özet

The particle-flow (PF) algorithm constructs a global description of each particle collision by producing a comprehensive list of final-state particles, and is central to event reconstruction in the CMS experiment at the CERN LHC. The existing PF implementation relies on physics-motivated heuristics and assumptions that can be replaced by machine-learning (ML) models trained directly on simulated data and naturally suited to modern graphics processing units (GPUs). A state-of-the-art ML-based PF (MLPF) reconstruction algorithm, implemented within the CMS software framework, is presented. The MLPF algorithm performs a learnable full-event reconstruction on GPUs, generalizes across detector conditions and collision energies, and replaces multiple modular reconstruction steps with a single unified model. Physics performance comparable to standard PF reconstruction is achieved in both simulation and data, with improved jet energy resolution and inference time. In simulated top quark-antiquark events under LHC Run-3 (2023–2024) conditions, the jet energy resolution improves by 10–20% for jets with transverse momentum between 30–100 GeV. Inference time is evaluated using simulated multijet events, with a median of 20ms per event on an Nvidia L4 GPU, compared to approximately 110ms for the standard CMS PF reconstruction.