Hardware

Universal Quantum posts decoder for transversal CNOT gates, reports lower simulated error rates

Trapped-ion developer Universal Quantum published a study (arXiv:2610.26784, Mark Webber and colleagues) on its Iterative Redecode Scheme (IRS), a decoder for transversal CNOT (tCNOT) gates. The article does not say the preprint has been peer reviewed, so it should be treated as not yet peer reviewed. IRS alternates local minimum-weight perfect matching with deterministic Pauli-frame propagation, which the authors say avoids decoding a global cross-block hypergraph. In simulations, the authors report about 4x lower logical error rates than the ghost protocol and about 15x lower than Logical Observable Matching on a deep random Clifford circuit (distance 11), and up to 25x lower than Reliable Logical Pauli-Products on a 3-block GHZ preparation (distance 19). For repeated logical S gates at distance 9, they report a circuit logical error rate of about 2.5 x 10^-5, roughly 4 to 10 times lower than LOM.

Why it mattersTransversal gates run in constant circuit depth with respect to code distance, but faults near them spread errors across code blocks and break standard pairwise matching decoders. Universal Quantum says IRS handles this with lightweight memory-style decoders and converges within a few iterations in simulation, though the results are simulated and come from the authors' own paper.

Read the full story at Quantum Computing Report →

Summarized by AI from the linked article.