Physics performance

Accurate identification of hadronic final states is crucial for harnessing the physics potential of collider experiments. At the FCC-ee, the pristine experimental environment, devoid of effects such as QCD ISR and PDFs, simplifies flavor tagging significantly compared to the (HL-)LHC, promising substantial improvements. Specifically, discriminating strange quark jets opens avenues for groundbreaking studies, including Z→ss production, rare Higgs boson decays, investigation of strange Yukawa coupling, and determination of CKM matrix elements through W decays.

We have implemented a multiclassifier neural network using a transformer-based architecture for flavour tagging at the FCC-ee. Using a transformer architecture means that the network is highly parallelisable during training, which is of utmost importance when evaluating different potential detector configurations. The network has achieved state-of-the-art strange quark discrimination and we will use it as a springboard to evaluate the feasibility of novel physics measurements and the required detector performance.

In addition to the flavour tagging development, we are also starting case studies that investigate the physics potential of FCC-ee, especially given above flavour tagger and vertex detector developments.