Study of low-momentum taus at CMS

Abstract

The tau lepton, which is the third-generation lepton, is the only lepton that decays hadronically as well as leptonically. Tau lepton reconstruction is thus challenging at a hadron collider such as the LHC. Standard tau reconstruction and identification at CMS typically requires tau leptons with transverse momentum (pT) above 20 GeV. However, several BSM models could benefit from the usage of lower pT tau leptons. For example, in the VLL singlet model, we are unable to significantly exclude vector-like taus of any mass. The selection of lower pT tau leptons would improve the acceptance of searches for the vector-like tau and thus could improve the sensitivity of the search. This is true of many other BSM searches as well. In this thesis, I will describe my study of the reconstruction and identification of taus at low momentum and improve the efficiency of these algorithms using machine learning models.