JEWEL 2.3 simulated pp and PbPb di-jet samples at 5.02 TeV - With and without Medium Response and Underlying Event

Samples of simulated di-jets generated with JEWEL 2.3 for both vacuum (pp) and medium (PbPb) collisions at \(\sqrt{s_{NN}} = 5.02\) TeV. First used in ``Quantifying vacuum-like jets in heavy-ion collisions: a Machine Learning study''. Two physics configurations are provided: Signal-Only (SO): jets produced without medium response and without underlying-event embedding. These represent the parton-shower modification in vacuum (pp) and in-medium (PbPb) without contamination and respective subtraction effects. Medium-Response + Underlying-Event (MR-UE): jets including JEWEL’s model of medium response with its dedicated subtraction prescription, embedded into a realistic heavy-ion underlying event which is subsequently subtracted using standard techniques. The corresponding pp baseline is embedded and subtracted in the same manner to isolate the effects of quark-gluon plasma (QGP) interactions from those due solely to underlying-event contamination. All events were produced at hadron level with a hard matrix-element transverse-momentum cutoff of \(p_T^{\text{min}} = 50\) GeV, and generation spectra were re-weighted by \(p_T^5\) to enhance high-\(p_T\) statistics. For the PbPb samples, the QGP medium follows the parametrization of Zapp et al., with \(\tau_i = 0.4\) fm/c,\(T_i = 590\) MeV, \(T_c = 170\) MeV, and \(0-10\%\) centrality, covering \(|\eta| < 4\). Jets were reconstructed using the anti-\(k_T\) algorithm (FastJet) with \(R=0.4\) from all particles within \(|\eta|<2.5\), keeping jets with \(p_T \in [80,230]\) GeV and at least two constituents. The resulting samples contain approximately \(1.7\times10^6\) jets for SO and \(1.9\times10^6\) jets for MR-UE, roughly equally divided between vacuum (pp) and medium (PbPb) classes. Each dataset includes two representations: Constituent four-momenta: for each jet, the set \(\{(\log(p_{T,i}/1,\text{GeV}), m_i, \Delta\eta_i,\Delta\phi_i)\}\), where \( \Delta\eta_i,\Delta\phi_i \) are defined relative to the jet axis. High-level observables: standard jet and jet substructure variables used in previous JEWEL + PYTHIA studies. These datasets support studies of machine-learning-based jet classification in heavy-ion collisions, particularly the effect of QGP-induced quenching and background contamination. Full methodological details are provided in the associated publication.