Revisiting the Electroweakino Sector of the Baryon Number Violating MSSM at the HL-LHC with Deep Neural Networks

We study the projected sensitivity of direct electroweakino production $pp \to \tildeχ_1^{\pm} \tildeχ_2^0$ at the HL-LHC in a simplified framework with wino-like, mass degenerate $\tildeχ_1^{\pm}$ and $\tildeχ_2^0$, and a bino-like lightest neutralino $\tildeχ_1^0$, assuming R-parity violating~(RPV) through the baryon number violating $λ^{\prime \prime}{112}u^c d^c d^c$ and $λ^{\prime \prime}{113}u^c d^c b^c$ operators. We consider three channels with the $λ^{\prime \prime}_{112}u^c d^c d^c$ RPV operator: $Wh$ mediated $1,\ell + 2,b + \rm E{!!!/}_T$, $Wh$ mediated $1,\ell + (\geq 2,j) + 2, γ+ \rm E{!!!/}_T$, and $WZ$ mediated $3\ell + (\geq 2 j) + \rm E{!!!/}_T$. In each channel, we train benchmark-specific multi-layer perceptrons (MLPs), analogous to signal-region classifiers, on the four-momenta of the final state particles along with a small set of higher-level observables to distinguish the signal from the dominant SM backgrounds. We find that the HL-LHC will be able to probe winos up to $\sim 900~$GeV, $\sim 780~$GeV, and $\sim 880~$GeV in the $Wh$ mediated $1,\ell + 2,b + \rm E{!!!/}T$, $Wh$ mediated $1,\ell + (\geq 2,j) + 2, γ+ \rm E{!!!/}T$, and $WZ$ mediated $3\ell + (\geq 2 j) + \rm E{!!!/}T$ channels, respectively, for $m{\tildeχ_1^0} \sim 50~$GeV, in the presence of $λ^{\prime \prime}{112}u^c d^c d^c$ couplings, at $2σ$ sensitivity. In case the $λ^{\prime \prime}{113}u^c d^c b^c$ operator is solely switched on, the projected sensitivity for winos reach up to $\sim 700~$GeV for $Wh$ mediated $1,\ell + (\geq 1,b), + (\geq 1j), + 2, γ+ \rm E{!!!/}_T$ and $\sim 850~$GeV for the $WZ$ mediated $3\ell + (\geq 1 b) + \rm E{!!!/}_T$ channel.

💡 Research Summary

This paper presents a comprehensive study projecting the sensitivity of the High-Luminosity LHC (HL-LHC) to directly produced electroweakinos (charginos and neutralinos) within a baryon number violating, R-parity violating (RPV) Minimal Supersymmetric Standard Model (MSSM). The analysis focuses on a simplified model framework featuring a wino-like, mass-degenerate chargino-neutralino pair (χ̃₁±, χ̃₂⁰) and a bino-like lightest neutralino (χ̃₁⁰). R-parity violation is introduced via the baryon number violating UDD-type superpotential operators, specifically λ’’{112}u^c d^c d^c and λ’’{113}u^c d^c b^c, which cause the otherwise stable LSP (χ̃₁⁰) to decay promptly into jets: uds for λ’’{112} and udb for λ’’{113}.

The study investigates five distinct final state channels arising from the decays χ̃₂⁰ → Z/h + χ̃₁⁰ and χ̃₁± → W± + χ̃₁⁰, followed by the leptonic decays of the W/Z bosons and Higgs decays to either b̄b or γγ. For the λ’’{112} coupling, three channels are analyzed: P1) Wh-mediated with h→b̄b, leading to 1 lepton + 2 b-jets + missing transverse energy (E̸T); P2) Wh-mediated with h→γγ, leading to 1 lepton + (≥2 jets) + 2 photons + E̸T; P3) WZ-mediated with Z→ℓℓ, leading to 3 leptons + (≥2 jets) + E̸T. For the λ’’{113} coupling, two channels are studied: P4) Wh-mediated with h→γγ, leading to 1 lepton + (≥1 b-jet) + (≥1 jet) + 2 photons + E̸T; P5) WZ-mediated with Z→ℓℓ, leading to 3 leptons + (≥1 b-jet) + E̸T.

The core of the analysis methodology is a multivariate technique employing benchmark-specific Multi-Layer Perceptrons (MLPs), analogous to optimized signal region classifiers. Unlike traditional cut-based analyses, these neural networks are trained on a rich set of input features, including the four-momenta of all final-state objects (leptons, jets, b-jets, photons) and a small set of higher-level kinematic observables such as the contransverse mass (MCT), angular separations (ΔR, Δφ), scalar sum of transverse momenta (HT), and a reconstructed top quark mass variable to suppress the dominant tt̄+jets background. This approach aims to capture subtle kinematic differences between the signal—characterized by high E̸T from the neutrino and boosted decay products due to large mass splittings—and the major Standard Model backgrounds (tt̄+jets, VV+jets, Zh+jets).

Event generation and detector simulation are performed using the MadGraph5_aMC@NLO, Pythia8, and Delphes3 software chain, assuming HL-LHC conditions (√s = 14 TeV, L = 3 ab⁻¹). Signal cross-sections are taken at NLO-NLL accuracy.

The key findings quantify the projected mass reach for the wino-like states. Assuming a lightest neutralino mass of ~50 GeV and a 2σ sensitivity, the HL-LHC can probe wino masses up to approximately 900 GeV, 780 GeV, and 880 GeV in the P1 (1ℓ+2b+E̸T), P2 (1ℓ+(≥2j)+2γ+E̸T), and P3 (3ℓ+(≥2j)+E̸T) channels, respectively, in the presence of the λ’’{112} coupling. If only the λ’’{113} operator is active, the projected sensitivity reaches up to ~700 GeV for the P4 channel (1ℓ+(≥1b)+(≥1j)+2γ+E̸T) and ~850 GeV for the P5 channel (3ℓ+(≥1b)+E̸T). The P1 channel offers the highest reach due to the large branching fraction of h→b̄b and efficient b-tagging.

The paper notes that the chosen couplings λ’’{112} and λ’’{113} are representative of broader classes of non-top-quark UDD couplings (e.g., λ’’{212}, λ’’{123}), and the results are generally applicable to those classes. Couplings involving top quarks (e.g., λ’’_{312}) require a separate dedicated study. Overall, this work demonstrates the significant potential of the HL-LHC, enhanced by modern multivariate analysis techniques, to explore the electroweakino sector in RPV SUSY scenarios, providing a detailed roadmap for future searches.