Reconstruct everything everywhere all at once: An end-to-end model for future colliders
Abstract
We present the Reconstruct Everything Model (REM), an end-to-end multitask model for sparse, multimodal sensor data, demonstrated in high-energy physics. To our knowledge, REM is the first demonstration of a single learned framework simultaneously reconstructing multiple particle types directly from hit-level measurements across the silicon tracker, calorimeters, and muon system in a full simulation. It performs hit assignment, particle identification, kinematic regression, and resonance reconstruction without externally reconstructed tracks or calorimeter clusters as inputs. REM improves particle-reconstruction efficiency and purity and jet-energy resolution relative to both conventional and machine-learned particle-flow approaches. In diboson events, the learned resonance reconstruction improves W/Z mass separation relative to conventional jet reconstruction. These results demonstrate a reconstruction chain from detector hits to particles to resonances, with potential applications to structured prediction in other large-scale, sparse, multimodal sensor systems.
Related
- Toponium Production at Future Muon Colliders and in heavy ion ultraperipheral collisions
- A phenomenological profile of Higgs portals at future colliders
- Probing Mis-Aligned Type-I Two-Higgs-Doublet Model in e+e- → ZH at Future Lepton Colliders
- Mono-Z signal for dark matter searches at future lepton colliders and cosmological interpretation
- Detecting photons in radiative Bhabha interaction with Si-strip and LYSO crystals at future e+e- colliders
BibTeX
@misc{hart2026reconstruct,
title = {{Reconstruct everything everywhere all at once: An end-to-end model for future colliders}},
author = {Max Hart and Kwok Yiu Wong and Samuel Van Stroud and Philippa Duckett and Tim Scanlon and Gabriel Facini},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.41889},
eprint = {2610.41889},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.41889},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Reconstruct everything everywhere all at once: An end-to-end model for future colliders
AU - Max Hart
AU - Kwok Yiu Wong
AU - Samuel Van Stroud
AU - Philippa Duckett
AU - Tim Scanlon
AU - Gabriel Facini
PY - 2026
DA - 2026/10/30
PB - arXiv
DO - 10.48550/arxiv.2610.41889
UR - https://arxiv.org/abs/2610.41889
AB - We present the Reconstruct Everything Model (REM), an end-to-end multitask model for sparse, multimodal sensor data, demonstrated in high-energy physics. To our knowledge, REM is the first demonstration of a single learned framework simultaneously reconstructing multiple particle types directly from hit-level measurements across the silicon tracker, calorimeters, and muon system in a full simulation. It performs hit assignment, particle identification, kinematic regression, and resonance reconstruction without externally reconstructed tracks or calorimeter clusters as inputs. REM improves particle-reconstruction efficiency and purity and jet-energy resolution relative to both conventional and machine-learned particle-flow approaches. In diboson events, the learned resonance reconstruction improves W/Z mass separation relative to conventional jet reconstruction. These results demonstrate a reconstruction chain from detector hits to particles to resonances, with potential applications to structured prediction in other large-scale, sparse, multimodal sensor systems.
ER -