From Qubits to Couplings: A Hybrid Quantum Machine Learning Framework for LHC Physics

Abstract In this paper, we propose an Hybrid Quantum Machine Learning (HyQML) framework to improve the sensitivity of double Higgs boson searches in the $HH \rightarrow b\bar{b}\gamma \gamma$ final state at $\sqrt{s} = 13.6~ \rm {\small TeV}$. The proposed model combines parameterized quantum circuits with a classical neural network meta-model, enabling event-level features to be embedded in a quantum feature space while maintaining the optimization stability of classical learning. In our benchmark setup, the hybrid model outperforms the pure quantum implementation, improves over XGBoost and perform similarly to the Graph Neural Network (GNN) baselines, achieving an expected 95% CL upper limit on the non-resonant double Higgs boson production cross-section of 1.9 × σSM and 2.1 × σSM under background normalization uncertainties of 10% and 50%, respectively. In addition, expected constraints on the Higgs boson self-coupling κλ and quartic vector-boson–Higgs coupling κ2V are improved relative to the pure quantum model and competitive with the classical baselines considered in this work.