⚛️ quantum physics

Pauli web of the $|Y\rangle$ state surface code injection

This paper utilizes ZX-calculus and the Pauli web formalism to analyze and understand the injection of the $|Y\rangle$ state on the rotated surface code.

Original authors: Kwok Ho Wan, Zhenghao Zhong, Ainhoa Zapirain

Published 2026-08-25

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Kwok Ho Wan, Zhenghao Zhong, Ainhoa Zapirain

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

In the quest to build a quantum computer that can solve problems beyond the reach of classical machines, scientists face a fundamental hurdle: keeping information safe from the slightest disturbance. Quantum bits, or qubits, are incredibly fragile, losing their delicate state if they interact too much with their environment. To overcome this, researchers use a method called error correction, which spreads a single piece of information across many physical qubits, creating a logical qubit that is robust enough to survive. However, to perform complex calculations, these logical qubits need to be fed special "resource" states, often called magic states, which act as the fuel for advanced operations. The challenge lies in injecting these states into the protective code without introducing the very errors the system is designed to prevent. This requires a precise sequence of steps to initialize the qubits and verify that the information has been transferred correctly.

A team of researchers has now used a visual language known as ZX-calculus to map out exactly how one specific type of state injection works. Instead of trying to inject the complex "magic" states used for the most powerful calculations, they focused on a simpler, more predictable state that behaves like a standard compass direction in the quantum world. By drawing the entire process as a diagram, they were able to trace the flow of information and show that the specific pattern of starting conditions used in previous proposals is necessary for the system to work. Their work confirms that when you set up the initial qubits in a very particular way—some starting in one state, others in a different state, arranged in triangular sections—the system naturally forms a continuous path of information that connects the beginning to the end, creating the correct logical state.

The researchers began by looking at a method proposed by Li and later adapted by Lao and Criger, which involves placing a special qubit in the corner of a grid and filling the rest of the grid with other qubits in specific starting positions. In their diagram, time flows from the bottom to the top. At the very bottom, they placed a single qubit in a specific state, while the qubits below a diagonal line were set to one type of state and those above the line were set to another. They then applied a series of measurements that check the relationships between neighboring qubits, a process that happens repeatedly to catch any mistakes. Using their visual tools, they traced how the properties of that single special qubit at the start spread out through the grid as these checks were performed.

What they found was that the specific arrangement of starting states creates a seamless chain of connections that links the input to the output. If the starting states were arranged differently, this chain would break, and the final result would be wrong. The diagrams showed that the qubits in the lower triangle and the diagonal line act as a foundation that supports the flow of information, while the qubits in the upper triangle provide the necessary structure to complete the circuit. This visual proof demonstrated that the seemingly odd pattern of starting conditions described in earlier papers is essential to ensure the logical Y correlator Pauli web terminates properly and recovers the correct logical state. It is like a bridge where every pillar must be placed in a specific spot; if even one pillar is moved, the bridge cannot span the gap.

A crucial part of their discovery involves how the system handles the very first round of checks. In a perfect world, the first measurements would always confirm that the qubits are in the right state. However, because of the way the initial states are mixed, some of these first checks produce random results that cannot be trusted. The researchers showed that there are specific areas of the grid where the first check is reliable and areas where it is not. The reliable checks are those where all the qubits involved started in the same state, allowing the system to verify the information immediately. The unreliable checks involve a mix of starting states, meaning the first measurement is essentially a guess. To fix this, the protocol requires the system to post-select, keeping only the outcomes where the first-round measurements on specific plaquettes yield a +1 value. This "post-selection" process ensures that the final logical state is clean and free from errors introduced during the injection.

The team also explored what happens if a mistake occurs right at the beginning, such as a qubit starting in the wrong state. Their diagrams showed how this error would travel through the system, flipping the results of specific measurements as it moved forward in time. By tracking these changes, they could see exactly how the system would identify and correct the error, provided the initial setup was correct. This level of clarity helps explain why the specific pattern of initialization is so important: it allows the system to distinguish between a genuine error and the natural randomness of the first measurement round. Without this clear path, the computer might mistake a random fluctuation for a real error and apply the wrong correction, ruining the calculation.

While this study focused on a simpler state to make the analysis possible, the researchers believe the same visual approach can be applied to the more complex "magic" states needed for full-scale quantum computing. They suspect that by adjusting the rules of their diagrams, they could map out the injection of those harder states as well. For now, their work provides a clear, visual confirmation of why the Li and Lao-Criger schemes work, turning a complex mathematical procedure into a understandable picture of how information flows through a quantum computer. This clarity is a vital step toward building machines that can reliably perform the calculations of the future.

Technical Summary: Pauli Web of the |Y⟩ State Surface Code Injection

Problem Statement
In fault-tolerant quantum computation, magic states (such as the $|T\rangle$ state) are essential resources. A significant challenge lies in encoding a bare physical magic state into a logical state on the surface code with high fidelity. While protocols exist for injecting $|T\rangle$ states (e.g., Li [5] and Lao-Criger [6]), understanding the underlying mechanics of these injection procedures, particularly the specific initialization patterns required, can be opaque. The authors aim to analyze the state injection procedure for the simpler $|Y\rangle$ state ( $|Y\rangle \propto |0\rangle + i|1\rangle$ ) on the rotated surface code. The $|Y\rangle$ state is chosen because it is a Clifford state, allowing for a complete end-to-end analysis using Pauli webs, unlike the non-Clifford $|T\rangle$ state.

Methodology
The paper employs ZX-calculus and Pauli webs (specifically following the coloring scheme of [10], which is opposite to that of [9]) to provide a diagrammatic understanding of the injection protocol.

Initial State Construction: The authors model the initialization of a distance $d=5$ $d = 5$ rotated surface code as proposed by Lao and Criger [6]. This involves a $d \times d$ $d \times d$ grid of physical qubits where:
- The top-left corner qubit is initialized in the $|Y\rangle$ state (a $\pi/2$ phase Z-spider).
- Qubits in the lower triangular and main diagonal regions are initialized in the $|+\rangle$ state.
- Qubits in the upper triangular regions are initialized in the $|0\rangle$ state.
Encoder Circuit: A single round of the surface code encoding circuit is applied, consisting of X-type followed by Z-type plaquette parity measurements.
Pauli Web Analysis: The authors apply Pauli web rules (Definition 2.7 from [10]) to the combined ZX-diagram of the initial state and the encoder. They post-select on $+1$ parity measurement outcomes and trace the propagation of "legs" (decorated edges) through the circuit to determine the resulting logical correlators.

Key Contributions and Results

Recovery of the Logical Y Correlator: The primary result is the demonstration that the Pauli web generated by the Li/Lao-Criger injection scheme, when applied to the specific $|Y\rangle$ initialization pattern, successfully recovers the logical $Y$ correlator of the rotated surface code. The authors show that the logical $Y$ correlator is the combination of the logical $X$ and logical $Z$ correlators (since $iY = ZX$ ).
Explanation of Initialization Patterns: The diagrammatic analysis clarifies the necessity of the specific "strange yet structured" initialization pattern (triangular separation of $|+\rangle$ and $|0\rangle$ states). The authors demonstrate that if the states within the highlighted triangular regions were altered or swapped, the logical $Y$ correlator Pauli web would fail to terminate properly, resulting in an incorrect logical state.
Post-Selection Mechanism: The paper provides a detailed explanation of why post-selection is required in the first round of parity measurements.
- Valid Plaquettes: Only 10 out of 24 plaquettes in a distance-5 code have Pauli webs that terminate on the initial states (6 X-type plaquettes with all $|+\rangle$ inputs and 4 Z-type plaquettes with all $|0\rangle$ inputs). These yield deterministic $+1$ outcomes in the absence of errors and are the only ones suitable for syndrome extraction in the first round.
- Invalid Plaquettes: The remaining 14 plaquettes span the boundary between the $|+\rangle$ and $|0\rangle$ regions. Their Pauli webs do not reach back to the initial time slice; instead, they project a new stabilizer into the future. Consequently, their first-round measurement outcomes are random ( $\pm 1$ with equal probability) and cannot be used for error syndrome inference. Using them could lead to misidentifying initialization errors (e.g., confusing an error on qubit 14 with one on qubit 13).
- Conclusion: The authors confirm that post-selecting on $+1$ outcomes for the 10 valid plaquettes is necessary to ensure high-fidelity logical state injection and correct Pauli-frame updates.

Significance
The paper claims that the Pauli web formalism offers a powerful, diagrammatic tool to visualize and verify complex state injection protocols. By applying this method to the $|Y\rangle$ state, the authors provide a rigorous proof that the Li/Lao-Criger schemes correctly generate the intended logical correlator. Furthermore, the analysis elucidates the structural reasons behind the specific initialization requirements and the necessity of post-selection, which were previously less transparent. The authors suggest that this approach could potentially be extended to the more complex $|T\rangle$ state injection by modifying Pauli web rules to accommodate $\pi/4$ phases, though this is presented as a future direction rather than a result of the current work.

More like this