GHZ-Preserving Gates and Optimized Distillation Circuits
Mingyuan Wang, Guus Avis, Stefan Krastanov · 2025
Read GHZ-Preserving Gates and Optimized Distillation Circuits by Mingyuan Wang, Guus Avis, Stefan Krastanov (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2510-25854.txt.
Introduction
Mingyuan Wang, Guus Avis, and Stefan Krastanov speed up the design of circuits that clean up noisy quantum states shared among many parties. The states are GHZ states, a standard form of multi party entanglement, which means the parties share linked quantum outcomes. Simulating such circuits in the usual way grows very costly as qubits are added.
Wang and the team give a listing and simulation method whose cost for these special gates stays constant, rather than growing with system size. Fast simulation lets them search for distillation circuits, which purify noisy states, that beat prior designs by a wide margin. The method also extends to related graph states and can be tuned to a chosen noise model.
Catalog record
| CREATOR | Mingyuan Wang, Guus Avis, Stefan Krastanov |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 84 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2510-25854
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
GHZ-Preserving Gates and Optimized Distillation Circuits Mingyuan Wang, Guus Avis, Stefan Krastanov Abstract Greenberger-Horne-Zeilinger (GHZ) states play a central role in quantum computing and communication protocols, as a typical multipartite entanglement resource. This work introduces an efficient enumeration and simulation method for circuits that preserve and distill noisy GHZ states, significantly reducing the simulation complexity of a gate on n n qubits, from exponential O ( 2 n ) O(2^{n}) for standard state-vector methods or O ( n ) O(n) for Clifford circuits, to a constant O ( 1 ) O(1) for the method presented here. This method has profound implications for the design of quantum networks, where preservation and purification of entanglement with minimal resource overhead is critical. In particular, we demonstrate the use of the new method in an optimization procedure enabled by the fast simulation, that discovers GHZ distillation circuits far outperforming the state of the art. Fine-tuning to arbitrary noise models is possible as well. We also show that the method naturally extends to graph states that are local Clifford equivalent to GHZ states. I Introduction In quantum information theory, multipartite entanglement is a key resource for a wide range of quantum protocols, from communication and cryptography to distributed quantum computing. Among the most studied forms of multipartite entanglement are Greenberger-Horne-Zeilinger (GHZ) states, which can be represented for n qubits as |\text{GHZ}\rangle=\frac{1}{\sqrt{2}}(|0\rangle^{\otimes n}+|1\rangle^{\otimes n}). (1) GHZ states are particularly valuable due to their characteristic non-classical correlations across multiple qubits, which play a central role in fundamental quantum tasks such as quantum secret sharing Hillery et al. 1999 , quantum conference key agreement Murta et al. 2020 ; Grasselli et al. 2022 ; Hahn et al. 2020 , error correction Bell et al. 2014 , and distributed quantum computing Yimsiriwattana and Jr 2004 ; Zhong et al. 2021 . However, their practical use is limited by the fragility of their entanglement under noise and imperfect local gates. This limitation motivates the use of entanglement distillation protocols, which aim to recover high-fidelity entangled states from multiple noisy copies via local operations and classical communication (LOCC) Bennett et al. 1996a ; Bennett et al. 1996b ; Dür and Briegel 2007 ; Acín et al. 2000 . The computational complexity of simulating entanglement distillation and error correction can be a bottleneck to large-scale circuit optimization. The exponential cost O(2^{n}) of state-vector simulation of n qubits can easily be avoided thanks to the more efficient Clifford-circuits formalism and its O(n) scaling for gate application Aaronson and Gottesman 2004 ; Gottesman 1998 . However, even O(n) can become prohibitive if it is employed in the inner loop of other algorithms, like circuit optimizers over large systems, or when real-time processing is required. This highlights the need for more efficient methods to simulate GHZ states. Another major obstacle lies in the vastness of the Clifford-circuit design space itself. Even though Clifford operations admit efficient simulation, the number of distinct Clifford circuits grows superexponentially with the number of qubits and gate layers, making exhaustive or even heuristic search intractable. This severely limits our ability to discover optimal or near-optimal GHZ-distillation protocols, especially when combined with noisy inputs and real-world hardware limitations. To address both challenges, we introduce a new framework based on the classification of GHZ-preserving operations—those local Clifford gates that map GHZ-basis states to other GHZ-basis states (terms defined in the following section). This restriction dramatically reduces the search space, as it excludes circuits that cannot possibly contribute to valid distillation. We then analyze the algebraic structure of the GHZ-preserving gate set and show that it forms a finite group that factorizes into a few small easy-to-enumerate subgroups. This group structure enables exact enumeration of allowed circuits and allows us to simulate their action in constant O(1) time by tracking permutations over a discrete GHZ basis. This advance enables efficient simulations even for large systems, making it feasible to model large networks and to employ computationally expensive circuit-optimization techniques. Our framework generalizes similar techniques used for Bell pair distillation Addala et al. 2023 ; Krastanov et al. 2019 ; Dehaene et al. 2003 . As a showcase of these algorithmic improvements, we use our new efficient simulation method together with a circuit optimizer based on genetic algorithms and simulated annealing to optimize GHZ-distillation circuits, while modeling the effects of network and gate noise. The efficient simulation enabled by our framework allows us to explore a much larger number of potential circuit designs, leading to more comprehensive optimization, and the circuits we produce significantly outperform anything else available in the literature. This is particularly relevant for quantum networks, where high-fidelity entanglement is essential for tasks such as distributed quantum computation and quantum conference key agreement. Furthermore, we extend our method to other quantum states, such as star-shaped and complete graph states, which are local Clifford equivalent to GHZ states. The remainder of this paper is organized as follows: In Section 2, we define the GHZ-preserving group of gates and present a group decomposition that enables efficient enumeration and gate simulation. We assume familiarity with the stabilizer formalism in this paper. Section 3 reviews current GHZ-distillation techniques, discusses noise modeling, and presents our superior distillation circuits, as well as the optimization techniques that generated them. In Section 4, we apply our decomposition to star-shaped and complete graph states, and we conclude in Section 5 with a discussion of implications to networking and other domains and future research directions. Detailed proofs of our group-decomposition theorems are provided in the Appendix B . II GHZ-preserving Group and Circuit Decomposition In this section, we formally define what it means for a unitary operation to preserve GHZ states and categorize all such operations. Informally, the set of “GHZ-preserving” operations is the set of gates that enable the distillation of such states (assuming noise is of the Pauli type, which is typical). We introduce both a fast simulation method for the application of such gates, and a compact enumeration for these gates. In later sections we will use these two capabilities to run circuit-optimization algorithms that find GHZ-distillation circuits much better than the state of the art. At first, we focus on two copies of an n -qubit GHZ state, each distributed among n nodes. We will prove that any GHZ-preserving transformation can be built out of Pauli operations, homogeneous unitaries consisting of the same gate applied to all nodes, (the H group ), and bilocal unitaries consisting of the same gate applied to two nodes (the B group ). This decomposition greatly simplifies the analysis and simulation of GHZ-based protocols by reducing the computational complexity and enumerating the set of useful gates. II.1 Decomposition of the GHZ-preserving Group Consider two identical n -qubit GHZ states, each shared among n nodes. Concretely, each node i holds two qubits in total, one from each GHZ state. Definition 1 (GHZ basis) . The n -qubit GHZ state is \lvert\mathrm{GHZ}\rangle=\;\frac{1}{\sqrt{2}}\Bigl(\lvert 0\rangle^{\otimes n}+\lvert 1\rangle^{\otimes n}\Bigr). (2) Any n -qubit state obtained from \lvert\mathrm{GHZ}\rangle by local Pauli operations (and hence sharing the same stabilizer generating set, up to differences in sign) is said to be in the GHZ basis . Concretely, this basis of 2^{n} states can be written as \left\{\,\Bigl(~\!\!\bigotimes_{i=1}^{n}P_{i}\Bigr)\,\lvert\mathrm{GHZ}\rangle\;\bigm|\;P_{i}\in\{I,X,Y,Z\}\right\}. (3)