articleNature CommunicationsJan 14, 2026GOLD OA

Transforming jet flavour tagging at ATLAS

GAG. AadEAE. AakvaagBAB. AbbottSASara AbdelhameedKAKarin Abeling

Centre National de la Recherche Scientifique · Aix-Marseille Université · +257 more institutions

PubMed
Indexed incrossrefdoajpubmed

Abstract

Jet flavour tagging enables the identification of jets originating from heavy-flavour quarks in proton-proton collisions at the Large Hadron Collider, playing a critical role in its physics programmes. This paper presents GN2, a transformer-based flavour tagging algorithm deployed by the ATLAS Collaboration that represents a different methodology compared to previous approaches. Designed to classify jets based on the flavour of their constituent particles, GN2 processes low-level tracking information in an end-to-end architecture and incorporates physics-informed auxiliary training objectives to enhance both interpretability and performance. Its performance is validated in both simulation and collision data.…

Citation impact

6
total citations
FWCI
85.75
Percentile
100%
References
80
Too recent for citation history.

Authors

2938

Topics & keywords

Keywords
  • Atlas (anatomy)
  • Flavour
  • Higgs boson
  • Large Hadron Collider
  • Interpretability
  • Quark
  • Jet (fluid)
  • Collision
No related works found for this paper.

Funding