Cell-level Transformers classify collimated ALP photon-jets versus single photons with AUC 0.98 and regress diphoton mass to ~64 MeV, beating shower-shape and other ML baselines in an ATLAS-like GEANT4 simulation.
Interpreting the 750 GeV Di-photon Resonance using photon-jets in Hidden-Valley-like models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
Motivated by the di-photon resonance recently reported by the ATLAS and CMS collaborations at $\sqrt{s} =13$ TeV, we interpret the resonance as a scalar boson $X (750)$ in hidden-valley-like models. The scalar boson $X$ can mix with the standard model Higgs boson and thus can be produced via gluon fusion. It then decays into a pair of very light hidden particles $Y$ of sub-GeV, each of which in turn decays to a pair of collimated $ \pi^{0} $'s, and these two $ \pi^{0} $'s decay into photons which then form photon-jets. A photon-jet ($ \gamma $-jet) is a special feature that consists of a cluster of collinear photons from the decay of a fast moving light particle (sub-GeV). Because these photons inside the photon-jet are so collimated that it cannot be distinguished from a single photon, and so in the final state of the decay of $X(750)$ a pair of photon-jets look like a pair of single photons, which the experimentalists observed and formed the 750 GeV di-photon resonance. Prospects for the LHC Run-2 about other new and testable features are also discussed.
fields
hep-ph 1years
2026 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
Transformer-based machine learning using low-level calorimeter signals for collimated photon identification at collider experiments
Cell-level Transformers classify collimated ALP photon-jets versus single photons with AUC 0.98 and regress diphoton mass to ~64 MeV, beating shower-shape and other ML baselines in an ATLAS-like GEANT4 simulation.