Flagging the Clifford hierarchy:~Fault-tolerant logical $\fracπ{2^l}$ rotations via measuring circuit gauge operators of non-Cliffords

Authors: Shival Dasu, Ben Criger Date: 2026-03-25 Paper ID: arxiv:2603.24573

Summary

This paper introduces a family of recursively defined flag circuits designed to detect logical errors induced by non-fault-tolerant $R_{\overline{Z}}(\fracπ{2^l})$ gates on distance-two CSS codes. The authors leverage these circuits to construct fault-tolerant logical $R_{Z}(\fracπ{2^l})$ and $R_{ZZ}(\fracπ{2^l})$ gates on iceberg codes using $O(l)$ gates and ancillas. Furthermore, they show an efficient method to prepare resource states for $R_{\overline{Z}}(\fracπ{2^l})$ via gate teleportation, bypassing high synthesis overheads for small $l$. The work also explores extending these techniques to arbitrary binary-digit phase rotations and demonstrates pathways to increase the fault distance of the constructions to 3 and 4 through code-specific enhancements and concatenation.

Key Contributions

• Developed a recursively defined sequence of flag circuits to detect logical errors from non-fault-tolerant $R_{\overline{Z}}(\fracπ{2^l})$ gates on distance-two CSS codes.
• Provided $O(l)$ gate and ancilla circuits for implementing fault-tolerant logical $R_{Z}(\fracπ{2^l})$ or $R_{ZZ}(\fracπ{2^l})$ gates on $k + 2, k, 2$ iceberg codes.
• Constructed $O(l)$ size fault-tolerant circuits for preparing $|\fracπ{2^l}\rangle$ resource states in the $7,1,3$ code, enabling low-overhead $R_{\overline{Z}}(\fracπ{2^l})$ rotations via gate teleportation for small $l$.
• Demonstrated methods to increase fault distance to 3 (Steane code, T gate) and to 4 (iceberg code concatenation), yielding a distance-4 logical $R_{\overline{Z}}(\fracπ{2})$ gate.

Limitations

The efficiency of the proposed circuits, particularly the $O(l)$ gate count and resource state preparation, is most advantageous when $l$ is small relative to the required precision for standard gate synthesis methods.

Limitations

The efficiency of the proposed circuits, particularly the $O(l)$ gate count and resource state preparation, is most advantageous when $l$ is small relative to the required precision for standard gate synthesis methods.

Links

• ArXiv Abstract
• PDF