Benchmarking Topological Qubit Stability Across 905 Circuits

Abstract

Recent advances in convolutional approaches have opened new routes to studying topological qubit stability. We introduce a convolutional framework and benchmark it against established baselines across 3619 trials. Our measurements agree with theoretical predictions to within 12%, resolving an open discrepancy in the literature. These findings suggest a practical path toward broader adoption across heterogeneous samples.