2026-05-29
14:37 |
Registre complet
|
2026-05-29
14:37 |
Registre complet
|
2026-05-29
14:37 |
GNN-based end-to-end reconstruction in the CMS Phase-2 High-Granularity Calorimeter
/CMS Collaboration
In this note, the first application of a one-pass, Machine Learning (ML) based imaging calorimeter reconstruction approach to the latest full CMS High-Granularity Calorimeter (HGCAL) simulation is presented.
The model is a Graph Neural Network (GNN) that directly processes the hits in the High
Granularity Calorimeter (HGCAL), one of the most important upgrades of the Compact Muon
Solenoid detector in preparation for the High-Luminosity phase of the Large Hadron Collider
planned to begin operations in 2030.
The network is trained to group hits originating from the same incident particle by assigning them to a common cluster. The accuracy of the reconstruction is evaluated through physics-inspired metrics that quantify how accurately the properties of individual particles are measured.
The algorithm is studied using simulations of different particle types in HGCAL and its performance is tested in single-particle environments.
While this exploratory study uses zero pile-up simulations and thus does not fully reflect expected detector performance, the results shown in this note demonstrate the viability of an end-to-end, ML-based reconstruction framework for the HGCAL..
CMS-DP-2026-057; CERN-CMS-DP-2026-057.-
Geneva : CERN, 2026 - 17 p.
Fulltext: PDF;
|
Registre complet
|
2026-05-29
14:37 |
HGCal Silicon modules at different absorbed doses
/CMS Collaboration
Several set of modules have been irradiated to study the hybrid assembly approach transfer tape + epoxy for the HGCal silicon modules.
In these slides, modules with different shapes, baseplate materials, sensor thicknesses and density, HGCROC readout electronic version (due to availability), have been built at University of California Santa Barbara (UCSB) and National Taiwan University (NTU), then they have been irradiated to different absorbed doses in a Co-60 facility in Croatia to study robustness of the assembly method at different expected level of absorbed dose. After irradiation, all modules have been measured and thermal cycled at UCSB. [...]
CMS-DP-2026-056; CERN-CMS-DP-2026-056.-
Geneva : CERN, 2026 - 25 p.
Fulltext: PDF;
|
Registre complet
|
2026-05-29
14:37 |
Results from HGCAL beam tests of 2025 for the scintillator setup
/CMS Collaboration
As part of its HL-LHC upgrade program, the CMS experiment is developing a High
Granularity Calorimeter (HGCAL) to replace the existing endcap calorimeters. The
HGCAL will be realized as a sampling calorimeter equipped with modules of silicon or
scintillator tiles read out by SiPMs. [...]
CMS-DP-2026-055; CERN-CMS-DP-2026-055.-
Geneva : CERN, 2026 - 15 p.
Fulltext: PDF;
|
Registre complet
|
2026-05-29
14:37 |
Results from HGCAL beam tests of 2025 for the silicon setup
/CMS Collaboration
As part of its HL-LHC upgrade program, the CMS experiment is developing a High
Granularity Calorimeter (HGCAL) to replace the existing endcap calorimeters. The HGCAL
will be realized as a sampling calorimeter equipped with modules of silicon or scintillator tiles
read out by SiPMs. [...]
CMS-DP-2026-054; CERN-CMS-DP-2026-054.-
Geneva : CERN, 2026 - 27 p.
Fulltext: PDF;
|
Registre complet
|
2026-05-29
14:37 |
Physics performance studies of the HGCAL TPG Stage2 Semi-emulator
/CMS Collaboration
This note presents recent developments and performance studies of the HGCAL Trigger Primitive Generation (TPG) system, with a primary focus on the "semi-emulator", a recent development of a cluster finder algorithm for the Stage2 of the TPG. This new development is a first implementation of an algorithm similar to that expected in the Stage2 FPGA firmware. [...]
CMS-DP-2026-053; CERN-CMS-DP-2026-053.-
Geneva : CERN, 2026 - 9 p.
Fulltext: PDF;
|
Registre complet
|
2026-05-26
11:43 |
Developments in Combine
/CMS Collaboration
This note describes recent developments in Combine, the CMS statistical analysis and combination tool, with a focus on the development of automatic differentiation support within the tool. In recent years, the RooFit team have added various new backends to RooFit, one of which is the codegen backend which enables AD via the
Clad tool. [...]
CMS-DP-2026-052; CERN-CMS-DP-2026-052.-
Geneva : CERN, 2026 - 16 p.
Fulltext: PDF;
|
Registre complet
|
2026-05-26
11:43 |
CMS Analysis Frameworks
/CMS Collaboration
One of the main challenges currently facing high energy particle physicists analyzing data from the Large Hadron Collider (LHC) at CERN is the unprecedented volume of both real data and simulated data that must be processed. This challenge is expected to intensify as the LHC enters its high luminosity phase, during which it is projected to deliver up to ten times more data than before. [...]
CMS-DP-2026-051; CERN-CMS-DP-2026-051.-
Geneva : CERN, 2026 - 20 p.
Fulltext: PDF;
|
Registre complet
|
2026-05-26
11:43 |
Differentiable Setup for a Top-Higgs Analysis
/CMS Collaboration
Autodifferentiation allows for the efficient computation of gradients in a complex computational graph. If the
analysis workflow, including the likelihood function, is implemented in a fully differentiable manner, the
parameters of an analysis can be optimized using the information of the statistical analysis.
In this work, we present how histogramming and the construction of the likelihood function can be
implemented in a differentiable way. [...]
CMS-DP-2026-050; CERN-CMS-DP-2026-050.-
Geneva : CERN, 2026 - 14 p.
Fulltext: PDF;
|
Registre complet
|