In simplified term “negatively interfere” means that less \(\tau\) leptons appeared (e.g. are counted) in the dataset as was expected from the Standard Model prediction.
There is a lot going into simply counting particles that would be beyond the scope of this project description. All details of the full analysis are in Ch. 3 of [1] and the paper is sheduled for publication in 2025.
I developed a \(\texttt{Python}\) framework for fitting the maximum likelihood estimator of the leptoquark signal strength with the statistical framework \(\texttt{cabinetry}\) [3] and determining the sensitivity to leptoquarks in absence of unblinded data. I carried out my work during 2022-2023 and the fitting framework is still (with some modifications of course) used today.
My contribution had a direct input to the background event generation, so events without leptoquark interference, and application of the kinematic cuts of possible signal events. The framework also provided a way to efficiently produce statistical results and provide them to the analysis group for rapid development of the analysis strategy.