arXiv:2609.31541
Sipser-Spielman meets Dijkgraaf-Witten: non-Abelian qLDPC codes via twisted sheaf gauge theory and almost-constant-overhead magic state fountain
Guanyu Zhu, Shi Jie Samuel Tan, Ryohei Kobayashi, Po-Shen Hsin
Abstract
Sipser-Spielman codes and Dijkgraaf-Witten twisted gauge theories are among the most profound ideas in the last few decades in the areas of computer science and mathematical physics respectively. This work unifies them in the same framework using the language of sheaf cohomology, and produces a new framework of twisted sheaf gauge theories describing a class of quantum low-density parity-check (qLDPC) codes with non-Abelian $D_4$ topological order. The corresponding twisted Clifford-stabilizer qLDPC code can be obtained from gauging a new type of 0-form sub-complex symmetry from a qLDPC code defined on a general sheaf complex, and can be interpreted as a topological defect network of non-Abelian $D_4$ patches glued together with proper gapped interfaces. As an application, one can use this to realize a magic state fountain via the gauging measurement of the addressable logical CZ gates as 0-form subcomplex symmetries in a 2D hypergraph-product (HGP) sheaf code. This includes a scheme of subdividing an arbitrary constant-rate 2D HGP code with parameters $[[n,Θ(n), Ω(n^{1/2})]]$ into a quantum sheaf code which allows preparation of $Θ(n^{1/2})$ logical CZ magic states in parallel, equivalent to the recent geometric construction using the code-to-manifold mapping in (arXiv:2601.06736). Moreover, using the recent sheaf complex and algebraic code constructions by Golowich-Tamo-Zhu (arXiv:2609.27801) with parameters $[[n,Θ(n^{1-ε}), Ω(n^{(1-ε)/2})]]$ for arbitrary small $ε$, one can prepare $Θ(n^{1-ε})$ CZ magic states in parallel and hence achieve an almost-constant magic rate.
AI Summary
The paper asks whether non-Abelian qLDPC codes and parallel magic-state preparation can be constructed directly on two-dimensional code complexes, without transversal CCZ gates or a code-to-manifold mapping. It develops twisted sheaf gauge theory: classical Tanner codes become graph complexes with local coefficients, supplying cup products for a type-III Dijkgraaf–Witten twist. Gauging produces non-Pauli Clifford-stabilizer codes interpreted as networks of $D_4$ topological-order patches. A family of “0-form subcomplex symmetries” supplies addressable logical CZ operators. Measuring their local dressed-check factorizations, then ungauging, projects initialized logical pairs into magic states. From good classical LDPC codes, the construction gives protected subsystem parameters $[[n,\Theta(n),\Theta(\sqrt n)]]$ and $\Theta(\sqrt n)$ disjoint magic-state outputs per run. Using recent classical codes with a local multiplication property, it obtains, for every fixed $\epsilon>0$, $k,s\ge n^{1-\epsilon}$ and $d=\Omega(n^{(1-\epsilon)/2})$, with constant check weight over a constant-size field. A constant fraction of pairs succeeds independently; characteristic-two constructions translate to qubits with constant-factor costs. Gauge invariance follows from cup-product identities. Distance is supported by direct Knill–Laflamme arguments, independently of a separate physical assumption about topological operators. The high-yield proof applies to gauge-fixed outcome sectors, not a protected–gauge factorization of the full twisted code. “Two-dimensional” refers to the complex, not planar-local connectivity. Preparation requires $O(d)$ measurement rounds, so almost-constant rate describes spatial yield, not spacetime overhead. Decoding and noisy-measurement thresholds remain unresolved, as do constant magic rate and linear-distance twisted codes. The alphabet size grows as $\epsilon\to0$.
Claude and ChatGPT were used to polish the writing and improve the structure of the paper, and to find references.
Error found: The abstract omits that the advertised distances concern protected subsystem sectors or gauge-fixed subspaces, rather than an unqualified distance statement for the full twisted code. The claimed BF-to-flatness reduction on arbitrary sheaf complexes lacks the complementary-degree nondegenerate pairing and integration prescription needed to show that summing multiplier fields enforces pointwise flatness.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 68 pages of main text and 17 figures + appendices