Validation and calibration of quantum hardware through the many-body quantum Mpemba effect

As quantum hardware diversifies, standardizing validation and calibration across different platforms (e.g., neutral-atom arrays, superconducting circuits) remains a significant challenge. Traditional methods like quantum state tomography scale poorly with system size, while randomized benchmarking often relies on simplifying assumptions about noise. This paper asks whether nonequilibrium many-body dynamics—specifically the quantum Mpemba effect—can be harnessed as a diagnostic tool to benchmark and calibrate quantum simulators more efficiently.

The researchers introduce a validation primitive that inserts a preprocessing sequence before a standard simulation task. By optimizing these preprocessing rotations, they prepare the system in states that either accelerate or decelerate the relaxation toward equilibrium. This creates a "fast" and "slow" pathway, allowing the researchers to probe the device's performance on the specific dynamical modes it is intended to simulate. They implemented this on two independent neutral-atom processors (QuEra's Aquila and Pasqal's Fresnel) using a six-atom Ising ring, comparing the measured relaxation trajectories against theoretical predictions to identify systematic errors and hardware-specific imperfections.

The study successfully demonstrates the many-body quantum Mpemba effect on two distinct hardware platforms, showing that the "fast" sequence reaches equilibrium significantly faster than the "slow" sequence, even when the latter starts closer to the equilibrium configuration. This provides a model-independent signature of the effect and a model-dependent benchmark of device quality. By analyzing the deviations between experimental data and theoretical models, the authors derive a figure of merit based on the Bhattacharyya coefficient, which serves as an upper bound on quantum fidelity. Furthermore, this approach provides a direct, actionable feedback loop for hardware developers to correct pulse sequences and improve the faithful reproduction of target dynamics.