Shedding light on dark matter at the LHC with machine learning

We investigate a weakly interacting massive particle dark-matter (DM) candidate in the form of a singlino-dominated lightest supersymmetric particle (LSP) within the Z 3 -symmetric next-to-minimal supersymmetric standard model (NMSSM). This framework gives rise to regions of parameter space where DM is obtained via coannihilation with nearby Higgsino-like electroweakinos and DM direct-detection signals are suppressed, the so-called “blind spots.” On the other hand, collider signatures remain promising due to enhanced radiative decay modes of Higgsinos into the singlino-dominated LSP and photons rather than into leptons or hadrons. Compared to MSSM scenarios with light bino- and winolike electroweakinos, the NMSSM allows for final states with multiple photons arising from cascade radiative decays, providing a distinctive collider signature. This motivates searches for radiatively decaying neutralinos; however, these signals face substantial background challenges, as the decay products are typically soft due to the small mass splits ( Δ m ) between the LSP and the Higgsino-like coannihilation partners. We apply a data-driven machine-learning analysis that improves sensitivity to these subtle signals, offering a powerful complement to traditional search strategies to discover a new physics scenario. Using an LHC integrated luminosity of 100 fb − 1 at 14 TeV, the method achieves a 5 σ discovery reach for Higgsino masses up to 225 GeV with Δ m ≲ 12 GeV and a 2 σ exclusion up to 285 GeV with Δ m ≲ 20 GeV . These results highlight the power of collider searches to probe DM candidates that remain hidden from current direct-detection experiments and provide a motivation for a search by the LHC collaborations using machine-learning methods.