Researchers have experimentally studied the impact of injecting coherent versus stochastic errors on the logical qubit performance using a bit-flip repetition code. The study, conducted on a transmon quantum processor, aimed to understand how different types of physical noise affect quantum error correction (QEC), a crucial component for building fault-tolerant quantum computers. Experimental results were compared with simulations adapted from a scalable free-fermion simulator, modified to efficiently sample stochastic noise in the quantum circuit.

Contrary to theoretical and simulation predictions, the experiment did not observe the expected difference in logical fidelity between coherent and stochastic error injection, for both distance-3 and distance-5 repetition codes. Simulations had suggested that these two types of noise should have distinct effects on QEC code performance. This discrepancy suggests that current noise models in experimental quantum systems might need refinement.

One hypothesis put forth by the team to explain this divergence is the presence of small drifts in qubit frequencies. These drifts could introduce phase-coherent noise that effectively "stochastifies" the injected coherent errors, making them behave more similarly to stochastic errors. This work underscores the complexity of characterizing and mitigating noise in real quantum platforms and contributes to a deeper understanding of how coherent errors affect experimental QEC, an essential step for the development of quantum computing.