The CMS collaboration reports the first evidence for the electroweak production of a pair of Z bosons produced with two jets in proton–proton collisions at 13 TeV. The result comes from events where one Z decays to two charged leptons (electrons or muons) and the other decays to two neutrinos, which escape the detector and appear as large missing transverse momentum. Using 138 inverse femtobarns of data collected in 2016–2018, they measure a production rate (cross section) in a defined fiducial region of 0.37 +0.14 −0.12 (stat) ±0.06 (syst) femtobarns, consistent with the Standard Model prediction of 0.39 ±0.06 fb.

To pick out the rare electroweak signal, the analysis selects events with exactly two same‑flavor, opposite‑charge leptons whose mass matches a Z boson, large missing transverse momentum to capture the neutrinos, and at least two jets. The two leading jets are required to have a large invariant mass (mjj > 400 GeV) and a large separation in rapidity (|Δηjj| > 2.5). These jet requirements target the vector boson scattering (VBS) topology, where the incoming quarks radiate vector bosons that then scatter to make the Z pair. The team used simulated samples for signal and many backgrounds, supplemented by data control samples—such as three‑lepton events to constrain WZ background and photon+jets events to model Z+jets events that can fake missing momentum. A graph neural network was trained to distinguish the signal from backgrounds and its output was fitted to the data to extract the signal size.

This measurement matters because VBS processes probe the electroweak force in a regime where the Higgs boson’s role is important for keeping scattering probabilities well behaved at high energy. Measuring electroweak Z‑pair production tests that part of the Standard Model and also gives sensitivity to possible new physics that would alter interactions among four gauge bosons. The paper also translates the data into limits on so‑called anomalous quartic gauge couplings using a dimension‑8 effective field theory framework, which parameterizes potential deviations from the Standard Model.