A Golay logical-zero circuit uses 44 two-qubit gates in six layers
Source check: Source checked. Not independent scientific confirmation.
Finding: An AI-assisted search found a circuit that prepares the logical-zero state of the 23-qubit Golay code with 44 two-qubit gates and two-qubit depth six. The author checked it by stabilizer algebra and full-state simulation. A previously posted 44-gate circuit used seven two-qubit layers under its stated labeling convention.
Technology: Claude Sonnet 5.5; Infinity Machine research agent
Family Claude: profile · models page
Publication: Preprint · Test context: direct test. Catalog categories, not an evidence grade.
- AI contribution
- Claude Sonnet 5.5 wrote a search program and found the circuit starting from a published depth-eight seed supplied by the operator.
- Test and result
- The checks compare the output state with the target up to wire relabeling. The controlled-Z gates are equivalent to CNOTs up to free one-qubit Clifford gates in this comparison.
- Limitation / still unknown
- The result assumes unrestricted connectivity and no extra qubits. It does not establish a depth or gate-count optimum, fault-tolerant preparation, or hardware performance.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://zenodo.org/api/records/23509367 · https://arxiv.org/html/2610.35579v1 · https://arxiv.org/html/2603.17743v1 · https://github.com/qecirc/qecirc-website/issues/198
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: linked separately.
Review date: 21 Nov 2026. Review scope: Zenodo v1 author's circuit record; Infinity Machine §3.2 and Appendix A.4; earlier QuSynth table and September circuit.
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 2/5 · Validation 3/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.