arXiv:2609.30159
Non-Abelian sheaf quantum LDPC codes: good and magical
Zimu Li, Fuchuan Wei, Zhengyi Han, Zi-Wen Liu
Abstract
Non-Abelian quantum codes connect quantum error correction, phases of matter, and computational resources. In this work, we develop a general framework for constructing non-Abelian quantum low-density parity-check (qLDPC) codes by gauging sheaf codes via cup products and use it to establish code families with constant encoding rate and linear distance, simultaneously achieving optimal asymptotic scaling in both parameters. In particular, we characterize explicit logical representatives of the full gauged code space, and provide a fundamental treatment of non-Abelian code distance in the sense of Knill--Laflamme based on combining expansion with cleaning to establish protection against arbitrary errors. We further construct an almost-good family with entire code spaces exhibiting long-range magic. Moreover, we use linearly addressable cup product to directly generate linearly many encoded magic states from gauging measurements of Clifford operators. These results extend good qLDPC codes beyond the Pauli stabilizer setting, lay the foundation for pursuing the no low-energy trivial magic conjecture and shed light on non-Abelian quantum phases and orders beyond geometric locality.
AI Summary
The paper asks whether non-Abelian quantum LDPC codes can achieve constant encoding rate and linear distance, with protection against arbitrary errors across the full code space rather than only a selected subsystem. It constructs such codes by gauging CSS sheaf codes using cup products: sparse $X$ checks are dressed with controlled-phase gates, while $Z$ checks remain Pauli. For every fixed integer $r\geq2$, it establishes families with parameters $[\![N,\Theta(N),\Theta(N)]\!]$ using Pauli and $r$-qubit multi-controlled-$Z$ operators. The code space is characterized explicitly by phase-weighted superpositions over coboundary orbits. Its dimension counts tuples of logical cohomology classes satisfying cup-product constraints; polarized representatives and related counting arguments give exponential dimension. A Knill–Laflamme analysis establishes a distance bound $d_{\mathrm{gauge}}\geq c\epsilon d$, where $d$ is the minimum parent-code distance, $\epsilon$ is an appropriate expansion constant, and $c$ depends only on fixed local parameters. Expansion provides small fillings of low-weight errors, and cleaning removes logical representatives from their neighborhoods. Suitable parent-code constructions yield linear distance. A separate construction has constant rate, distance $\Theta(N/\mathrm{polylog}\,N)$, and long-range magic throughout its code space, certified through a dimension obstruction and covering-space induction. Addressable gauging measurements also yield $\Theta(N)$ separate encoded magic states with high probability. The paper does not establish long-range magic throughout a linear-distance code family or throughout a low-energy sector, and therefore does not resolve NLTM. Perturbative stability, quantitative robustness to noise and approximate preparation, efficient decoding, and fault-tolerant measurement costs remain open.
Central claim may not hold: The non-Abelian-lift extension used for linear distance does not preserve the explicitly defined product squares, and the paper does not fully specify and justify the left-right replacement used to obtain the headline parameters. The alternative good-qLTC construction changes a block to degree three but counts its logical classes using H^2 rather than H^3, leaving its displayed constant-rate argument unsupported. The stated Knill–Laflamme theorem incorrectly identifies a bound on the sum of X- and Z-component weights with a bound on physical Pauli weight.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.