SciPost Phys. Proc. 18, 021 (2026) ·
published 29 January 2026
|
· pdf
The Large Hadron Collider (LHC) offers a unique opportunity to investigate $\mathcal{CP}$ violation in the Yukawa coupling between the Higgs boson and the top quark by studying Higgs production in association with top quarks; this is of fundamental importance, seeing that the $\mathcal{CP}$ properties of the Higgs boson are yet to measure with high precision. To address this, the focus of this work has been an extension of the simplified template cross-section (STXS) framework, devised to be sensitive to $\mathcal{CP}$ effects. Our study focused on $\mathcal{CP}$-sensitive observables across multiple Higgs decay channels, comparing their performances. The result indicates that the most efficient extension of the current binning used in the STXS framework, which currently uses the Higgs boson's transverse momentum $p_{T,H}$, requires adding one further split using $\mathcal{CP}$-sensitive observables. Between these observables, one of the best is the Collins-Soper angle $|\cos\theta^*|$, a variable derived from momenta information of the top quarks. We have investigated the improvement brought by our two-dimensional STXS setup and compared it to the currently employed methodologies, finding an increase in performances at an integrated luminosity of $300 \mathrm{ fb}^{-1}$. Moreover, our results highlight that this advantage seems to be present also at $3000 \mathrm{ fb}^{-1}$.
SciPost Phys. 19, 155 (2025) ·
published 16 December 2025
|
· pdf
The ATLAS experiment at the Large Hadron Collider explores the use of modern neural networks for a multi-dimensional calibration of its calorimeter signal defined by clusters of topologically connected cells (topo-clusters). The Bayesian neural network (BNN) approach not only yields a continuous and smooth calibration function that improves performance relative to the standard calibration but also provides uncertainties on the calibrated energies for each topo-cluster. The results obtained by using a trained BNN are compared to the standard local hadronic calibration and to a calibration provided by training a deep neural network. The uncertainties predicted by the BNN are interpreted in the context of a fractional contribution to the systematic uncertainties of the trained calibration. They are also compared to uncertainty predictions obtained from an alternative estimator employing repulsive ensembles.
