arXiv:2610.13880
Bicycle Race: small two-block codes on the Pareto frontier
Gideon Lee
Abstract
Two-block group-algebra codes are among the simplest quantum LDPC codes, and many of them are translation invariant on a twisted torus, which suits neutral-atom arrays and other platforms with movable qubits. We ask how far they go at small sizes when searched systematically across all groups. We use algebraic constraints and screens to design a search pipeline capable of enumerating broad and complete swathes of two-block codes on a laptop. With it we census about $10^8$ two-block codes with up to $320$ qubits at check weights five to nine. Several of these censuses are exhaustive, and support complete statements; for instance, the Gross code is, up to equivalence, the only abelian weight-six two-block code with $n \le 200$ to reach $kd^2/n=12$, no weight-six two-block code with $n \le 300$ beats $[[288,16,16]]$, and the known $[[64,18,8]]$ is the best two-block code by $kd^2/n$ across weights five to nine with $n \le 94$. The censuses are also large enough to give insights into where good codes live - they concentrate in a handful of groups; abelian groups give the best codes at small $n$ and non-abelian groups at large $n$, with the crossover appearing to move out as the check weight grows; and high quality codes become $100$ to $1000$ times rarer per draw from weight six to weights eight and nine. Using the group algebra and the census data to steer a statistical search finds four to fifteen times as many good codes as uniform draws in equal time. Among the codes found are $[[102,20,10]]$ and $[[120,24,10]]$ at check weight nine, which we simulate in noisy syndrome-extraction circuits, and $[[186,20,18]]$, a weight-nine code with every check in a $5\times5$ window. At check weights seven to nine, our best codes exceed the best published $kd^2/n$ with exact distance at the same check weight in most length bands up to $320$ qubits. All codes are available in a public database.
AI Summary
The paper asks how competitive two-block group-algebra CSS qLDPC codes remain at small blocklength when searched across finite groups rather than restricted ansätze. It searches approximately $10^8$ candidates with $n\le320$ and check weights $5$–$9$, using equivalence reductions, algebraic screens, randomized logical witnesses, and exact connected-cluster distance calculations. A public database contains about $12{,}400$ codes and comparisons with $2{,}611$ literature records. Exhaustive censuses cover weights five and six through $n=300$, weights seven and eight through $n=100$, and weight nine through $n=94$. They establish that the Gross code is the unique abelian weight-six equivalence class attaining $kd^2/n=12$ for $n\le200$, and that no weight-six two-block code with $n\le300$ exceeds the $[[288,16,16]]$ code’s $14.22$. New examples include weight-nine $[[102,20,10]]$, $[[120,24,10]]$, and $[[186,20,18]]$, the latter with checks fitting a $5\times5$ twisted-torus window. New codes extend the reported CSS parameter frontiers at weights seven through nine. Group-algebra-component targeting increases good-code yield by factors of four to fifteen in equal-time comparisons. The contribution is a search pipeline and census, not a new construction or asymptotic bound. Higher-weight searches are incomplete, and observed group-structure trends remain partly unexplained. Distances above roughly fifteen primarily rely on the custom exact engine, with independent upper-bound checks. Circuit simulations cover only two codes, use circuit distance eight rather than ten, and omit idle and movement noise. Logical operations and hardware-level schedule comparisons are not evaluated.
The exception is App.~\ref{app:literature}, which concerns how the literature code set was constructed, which was AI-drafted and only human-edited for clarity.
Error found: The paper inconsistently reports the published weight-eight [[306,22,24]] code as having confirmed exact distance 24 in Section 3.1 but unresolved distance with only d ≥ 23 proved in Appendix E.1. The sampling explanation incorrectly treats every component of an abelian group algebra over F₂ as a field, overlooking non-simple components for even-order abelian groups where nonzero singular elements need not vanish.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 59 pages: 18 main text + 20 appendices + 8 references + 13 code tables and supplementary figures after the references. 17 figures, 11 tables; v2: corrected literature credits in Tables 3 and 9 Code database: https://github.com/gideonlsx/bicycle-race