November 3, 2026 -- Researchers at the University of Regensburg and their international partners have found a way to significantly reduce errors in quantum bits. In an experiment, they demonstrated that a single additional measurement before a series of quantum operations can cut the error rate by more than half. The results have been published in Applied Physics Reviews.

Quantum computers are widely regarded as a promising technology for solving problems beyond the reach of today’s computers. One of their main challenges, however, is that their fundamental building blocks, known as quantum bits are extremely sensitive to disturbances.

A quantum bit, usually shortened to qubit, is the basic unit of information in a quantum computer and plays a role similar to that of a bit in a conventional computer. While classical bits store information as either 0 or 1, qubits are realized by physical systems governed by the laws of quantum mechanics. In this study, the qubit consisted of a tiny superconducting electrical circuit on a chip that was controlled using microwave signals.

When a Qubit Goes Out of Tune

Even very small changes in a qubit’s environment can alter its properties and introduce errors. The researchers focused on a superconducting qubit whose so-called resonance frequency would uncontrollably switch between two known values.

The resonance frequency determines which microwave signals can be used to control the qubit with high precision. If the frequency changes, the control signals must be adjusted accordingly.

One way to think about this is as a musical instrument. If an instrument goes slightly out of tune, the same movements no longer produce exactly the intended notes. Similarly, when a qubit’s resonance frequency shifts, the control signals become less precise, and the likelihood of errors increases.

Importantly, these two frequencies do not represent different computational states of the qubit. Rather, they are two possible properties of the same physical device. The qubit remains capable of occupying the quantum states required for computation at either frequency.

The frequency switching is caused by interactions between the qubit and tiny defects in the material. Although the two possible frequencies were known, determining which one was active at a given moment and adjusting the control signals accordingly had remained a challenge.

A Rapid Frequency Check Before Computation

This is where the new method comes in. The researchers developed a technique that allows them to identify which frequency is active before performing a sequence of quantum operations.

To do this, they first prepare the qubit in a quantum state and allow it to evolve for a short period of time. The evolution depends on which resonance frequency is present. A final measurement then provides the information needed to distinguish between the two possibilities.

The result is immediately used to adjust the subsequent control signals. This approach is known as feedback, because information obtained from a measurement is fed directly back into the control of the system.

Previous studies had already shown that the two possible frequencies of such a qubit could be distinguished. The new work takes a crucial step further: the information is used in real time to optimize qubit control and reduce errors.

Error Rate Reduced by More Than Half

The experiments were carried out using a superconducting quantum circuit at the Israeli Quantum Computing Center. Measurement data were processed in real time, allowing the control signals to be updated immediately.

To evaluate the effectiveness of the method, the researchers compared the performance of the qubit with and without feedback. Without adjustment, the quality of the quantum operations fluctuated significantly over time. With feedback enabled, the qubit performed consistently with high fidelity.

The average error rate decreased from approximately 0.14 percent to 0.065 percent, corresponding to a reduction of about 54 percent. Compared with the largest error observed without feedback, the reduction reached 77 percent.

Toward More Reliable Quantum Computers

The method currently works under specific conditions. The possible frequencies must be known in advance and sufficiently distinct from one another. In addition, the frequency must remain stable long enough for the updated control signals to have an effect. Nevertheless, the study demonstrates that a single additional measurement can significantly improve the reliability of a qubit. 

“At first sight, this may seem like a proof-of-principle demonstration with limited practical value”, says Ferdinand Kuemmeth. In the long run, as quantum materials become cleaner and cleaner, millions of nearly perfect qubits will need to be operated simultaneously, some of which will likely interact with a few remaining material defects. “I am convinced that in that emerging era of quantum computing scaling, our methods will become very valuable and make a tangible impact on the actual computational power of quantum computers. In fact, we have already developed the methods needed to automatically and efficiently detect unknown qubit frequencies” says Kuemmeth. He is Professor of Experimental Physics at the University of Regensburg and an Affiliate Professor at the Niels Bohr Institute, University of Copenhagen and led the study. His research focuses on semiconductor and superconducting quantum devices for quantum information processing.

The researchers see this as a promising approach to mitigate the effects of unavoidable material defects in future quantum processors, to automate the stabilization of qubits, and thereby to facilitate the scaling of quantum computers.

“It is becoming increasingly clear that the commercial value of quantum computers relies on scaling today’s quantum computers to much larger qubit numbers, without compromising on qubit performance and control fidelity. That’s exactly why efficient, autonomous, real-time qubit control is becoming more important than ever before”, Fabrizio Berritta, who remotely operated the quantum computing hardware in the Israeli Quantum Computing Center from his office at Massachusetts Institute of Technology (MIT), adds. He is a Marie Skłodowska-Curie Postdoctoral Fellow hosted by the University of Regensburg through the QuARTEC project and currently based in MIT’s Engineering Quantum Systems group, working on adaptive quantum error correction and qubit calibration.

The study involved researchers from the University of Regensburg, Leiden University, and the Norwegian University of Science and Technology (NTNU). Industry partners included Quantum Machines and Qolab.