Google Yoked Surface Codes

News from the land of quantum error-handling strategy—realm of many puzzles and missing pieces.

‘Yoked’ codes are two-dimensional error-correcting codes that add a ‘yoke’, or row and column of simpler high-rate parity check codes, along the outer edge of an array of surface codes. These ‘yokes’ enable further error suppression in a quantum memory with fewer qubits than required with the surface code alone. In a new paper, Google researchers report a space savings by carefully laying out the surface code qubits and scheduling the circuits.

Recently, there has been increasing interest in using high-rate codes on top of the surface code to reduce the number of qubits in a quantum memory. For, example, this has been studied for quantum LDPC + surface code, and for light-weight error-detecting codes + surface code in this new paper. Leading researchers currently expect hierarchical coding approaches to work for more efficient LDPC codes too, like the bivariate bicycle codes, but this is future research.

Read the publication in full for a closer look at the work. Link below!

/https://www.nature.com/articles/s41467-025-59714-1