Answers are grounded in this paper and cite their sources.
Demonstrates improved signal reconstruction in high-energy collisions, suggesting enhanced analysis methods.
A study for the associated production of a Higgs boson with a top quark pair (tt̄H) in the fully hadronic final state is presented, focusing on the H → bb̄ decay channel using proton-protoncollision data collected by the CMS experiment during Run 3 in 2024. Resolving the jet-parton assignment task in this high jet multiplicity environment represents a major source of combinatorial ambiguity. To address this limitation, a multivariate approach based on Symmetry-Preserving Attention Networks (SPANeT) is implemented to optimize the event reconstruction. Evaluated on simulated signal samples, the network achieves a Higgs boson acceptance of 63% and an assignment efficiency of 57% in the inclusive baseline category. This marks a significant improvement by nearly a factor of three compared to the 21% efficiency obtained via traditional geometric matching methods based on angular proximity (∆R). Finally, the network’s susceptibility to systematic biases is validated using a pure tt̄ + bb background control sample. While the model successfully reconstructs the signal topology, a noticeable mass-sculpting effect is observed in the background spectrum. To mitigate this effect in future analysis workflows, an impartial training scheme utilizing multi-mass signal samples is proposed to decorrelate the network’s predictions from the target Higgs boson mass.
No takes yet. Share an insight, caveat, or question.
Ruth Macías Asenjo (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: