arXiv:2610.09341
Building codes with transversal CCZ using projective geometry and SAT solvers
Bohan Lu, Kenneth R. Brown
Abstract
We construct CSS codes with three logical qubits whose logical CCZ gate is implemented by transversal physical $T/T^\dagger$ gates and whose distance is at least 3. These codes start with 8-divisible $X$-stabilizers defined by projective geometry, which guarantees $d_Z \geq 3$. A SAT solver then finds three $X$-logical operators compatible with the transversal CCZ. We build codes at thirteen block lengths from $n=48$ to $n=496$, with $d_Z=3$ or 4, and illustrate the method with the [[48,3,3]] code $Q_{48}$. Additionally, we prove that no CSS code with three logical qubits, $Z$-distance at least 3, and a quasi-transversal CCZ gate (transversal $T$ with a diagonal Clifford correction) exists below $n=39$. The [[47,3,3]] code of Jacinto et al. attains the minimum length in this class, as established by the classification of Wills, Jain, and Singh.
AI Summary
The paper asks how to construct small CSS codes encoding three qubits whose logical CCZ is implemented directly by transversal physical $T/T^\dagger$ gates, and how short such codes can be. It distinguishes this implementation from quasi-transversal CCZ, which permits an additional diagonal Clifford correction containing entangling $CZ$ gates. The construction fixes a projective, $8$-divisible classical code for the $X$ stabilizers using complements of subspaces in binary projective geometry. Distinct nonzero stabilizer columns guarantee $d_Z\ge3$, while divisibility supplies the stabilizer-only Campbell–Howard overlap conditions. SAT searches or structured label maps supply three logical $X$ rows satisfying the remaining conditions. A separate binary feasibility problem modulo $4$ determines whether physical $T/T^\dagger$ signs eliminate the Clifford correction. The explicit $[[48,3,3]]$ example has $d_X=16$, $d_Z=3$, and six independent $X$ stabilizers; its logical CCZ uses $26T+22T^\dagger$. Codes are reported at thirteen block lengths from $48$ to $496$, with $d_Z=3$ or $4$. A lower bound excludes all three-logical-qubit CSS codes with $d_Z\ge3$ and quasi-transversal CCZ at $n\le38$, using projective-support and stabilizer-rank reductions, Reed–Muller low-weight classifications, and exhaustive logical-label tests. A separate concurrent classification establishes the quasi-transversal minimum of $47$; the cited $[[47,3,3]]$ code fails the paper’s sign-feasibility test. The $48$-qubit construction is presented as the smallest known single-qubit-transversal example. Most larger-code counts are only lower bounds; characterizing admissible supports and extending to other gates or more logical qubits remain open.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 40 pages, 3 figures. Code and data: https://github.com/jerrylvx/gate2code-public