Speaker
Description
Accurate identification of jets that originate from heavy-flavor hadrons is pivotal for many ATLAS analyses, from Higgs-boson and top-quark measurements to searches for new physics. We present the newest heavy-flavor taggers from ATLAS, which introduces a full-transformer architecture tailored to the environment of Run 2 and Run 3.
The flavor tagging transformer processes low-level track, vertex, neutral particle, and muon information to extract correlations between the inputs and infer the origin of the jet. Compared with the current Run 3 baseline, the new model achieves better separation of $b$- and $c$-jets from light-flavor jets across a wide kinematic phase-space.
In this talk, we will discuss the architecture and training workflow, as well as the newest results. Applications in boosted object tagging or trigger may also be discussed.