paper

New graph-neural-network flavor tagger for Belle II and measurement of in decays

arXiv:2402.17260 · doi:10.1103/PhysRevD.110.012001

Abstract

We present GFlaT, a new algorithm that uses a graph-neural-network to determine the flavor of neutral mesons produced in decays. It improves previous algorithms by using the information from all charged final-state particles and the relations between them. We evaluate its performance using decays to flavor-specific hadronic final states reconstructed in a 362 sample of electron-positron collisions collected at the resonance with the Belle II detector at the SuperKEKB collider. We achieve an effective tagging efficiency of , where the first uncertainty is statistical and the second systematic, which is better than the previous Belle II algorithm. Demonstrating the algorithm, we use decays to measure the mixing-induced and direct violation parameters, and .

15 pages, 9 figures

References in corpus (11)

Cited by in corpus (6)