January 30, 2026: Quantum computers hold the potential to solve problems far beyond the reach of our best classical supercomputers. However, this immense power comes with a critical vulnerability: quantum states are incredibly "fragile to noise." Even the slightest interaction with the environment can introduce errors, corrupting calculations and rendering the results useless. The essential solution to this problem is Quantum Error Correction (QEC), a sophisticated system for detecting and correcting these errors in real-time. At the heart of any QEC system is a component called the "decoder." Its job is to analyze error signals, called syndromes, and deduce the most likely error that occurred. This is a monumentally difficult task. Finding the optimal correction is an "NP-hard" problem, meaning the computational time required to solve it explodes as the number of qubits grow. As a result, exact decoders are far too slow for practical use. On the other side are fast, approximate decoders that rely on "heuristic" rules. While quick, their "limited" accuracy can fail to protect the computation, leading to a difficult trade-off between speed and performance. To bridge this gap, we have developed a family of hierarchical quantum decoders based on the lasserre sum-of-squares hierarchy, which transforms the decoding task into a sequence of semidefinite programs. This framework provides a tunable spectrum of solvers where higher levels progressively account for complex error correlations, allowing us to move from fast approximations to rigorous solutions with mathematical precision. A particularly nice feature of this approach is its ability to provide a formal certificate of optimality through a rank-loop condition, which mathematically confirms when the absolute best solution has been found. ArXiv link
January 29, 2026: Our work, “Contextuality of quantum error-correcting codes,” has been published in PRX Quantum. As highlighted in the popular summary, the paper introduces a new and practical “litmus test” for quantum engineers: any error-correcting scheme that lacks contextuality cannot support universal quantum computation. The work was also featured in Physics magazine. According to the American Physical Society, this is Physical Review’s highest level of editorial highlight, reserved for papers of exceptional interest and impact, and typically associated with higher citation rates than papers published in Science or Nature. Many thanks to all collaborators.
January 22, 2026: I have been invited to speak at a special Reviews of Modern Physics (RMP) session (March 18, 2026) at the 2026 APS Global Physics Summit in Denver. The title of my talk is "Noisy Intermediate-Scale Quantum Algorithms: Opportunities, Limitations, and Lessons."
January 3, 2026: Our paper “Contextuality of quantum error-correcting codes” has been selected for coverage as a Viewpoint in Physics Magazine by the American Physical Society. A Viewpoint is the Physical Review’s highest level of editorial highlight, written by an expert in the field and reserved for papers of exceptional interest and impact. Articles receiving this distinction are, on average, cited more frequently than physics papers published in Science or Nature. Both the paper and the accompanying Viewpoint are currently scheduled for publication in late January 2026.
December 17, 2025: Concluded an 18-session tutorial series on Quantum Error Correction and Fault Tolerance. The complete set of lecture recordings can be found here.