Source-linked AI summary
Top-tagging: A Method for Identifying Boosted Hadronic Tops
David E. Kaplan, Keith Rehermann, Matthew D. Schwartz, Brock Tweedie
TL;DR
Boosted hadronic tops can resemble single jets, making conventional identification difficult and all-hadronic t¯t resonance searches vulnerable to dijet backgrounds. The paper introduces large-radius clustering, iterative declustering, and top-decay kinematic cuts to identify such jets, with efficiencies and limitations evaluated in Monte Carlo studies.
Problem
Boosted top decay products can be collimated and difficult to resolve, while all-hadronic t¯t searches face large dijet backgrounds and high-pT b-tagging challenges.
Source: p. 1Method
The method captures potential substructure with a large-radius jet, iteratively declusters it into subjets, and applies top-mass, W-mass, and helicity-angle constraints.
Source: p. 2Results
The method’s tagging and mistagging efficiencies are evaluated for top, gluon, and light-quark jets, and top-tagging reduces the dijet sample to the level of the t¯t sample.
Source: pp. 3–4Takeaways & Limitations
Top-tagging can support all-hadronic t¯t resonance searches and applications involving semi-leptonic tops, missing energy, and high-pT b-tagging studies.
Source: p. 4The study relies on Monte Carlo modeling of t¯t and dijet distributions, whose correctness is an explicit concern.
Source: p. 3
Abstract
from arXiv · showhide
A method is introduced for distinguishing top jets (boosted, hadronically decaying top quarks) from light quark and gluon jets using jet substructure. The procedure involves parsing the jet cluster to resolve its subjets, and then imposing kinematic constraints. With this method, light quark or gluon jets with pT ~ 1 TeV can be rejected with an efficiency of around 99% while retaining up to 40% of top jets. This reduces the dijet background to heavy t-tbar resonances by a factor of ~10,000, thereby allowing resonance searches in t-tbar to be extended into the all-hadronic channel. In addition, top-tagging can be used in t-tbar events when one of the tops decays semi-leptonically, in events with missing energy, and in studies of b-tagging efficiency at high pT.