- Accepted Paper
Bias-tailored Clifford deformations for surface codes
Phys. Rev. A - Accepted 17 November, 2026
DOI: https://doi.org/10.1103/b14p-xpg1
Phys. Rev. A - Accepted 17 November, 2026
DOI: https://doi.org/10.1103/b14p-xpg1
Clifford deformations of surface codes provide a promising approach for designing quantum error-correcting codes tailored to biased Pauli noise. Variants of surface codes, such as the and codes, achieve improved threshold performance under -biased noise by deforming the stabilizers of the surface code through single-qubit Clifford operations. However, high threshold performance does not necessarily translate into low logical error rates at finite code distances over a range of bias values. Specifically, we consider the finite-bias regime of practical interest, ranging from to . Furthermore, identifying high-performing Clifford-deformed surface codes (CDSCs) remains challenging due to the large space of possible Clifford deformations. To address these challenges, we propose a practical construction of CDSCs tailored to biased noise by leveraging the topological structure of surface codes. In particular, larger-distance CDSCs are systematically generated from carefully designed small deformation patterns, referred to as , enabling scalable code construction. To design repeating units, we analyze the logical operator structures of CDSCs under biased noise and introduce a new metric for estimating logical error rates. Using the logical operator structures and the proposed metric, we identify repeating units for constructing high-performing CDSCs tailored to representative bias values , , and . Numerical results demonstrate that, under -biased noise with ideal syndrome extraction, the proposed CDSCs achieve lower logical error rates than existing CDSCs, including the , , and random CDSCs, while achieving thresholds near or exceeding the hashing bound.
If the author has provided any supplemental materials with this article they will be available upon publication of the version of record.