Pan and Colleagues Implement Patch-Based Logical Operations for Surface-Code Processing

A new set of tools for fault-tolerant logical operations brings practical quantum computation closer to reality. Weiping Lin and colleagues from University of Science and Technology of China, Tsinghua University and Zhongguancun Laboratory, have experimentally realised key elements of surface-code logical processing using a 107-qubit superconducting quantum processor. They implemented reusable primitives for manipulating surface-code patches, enabling logical state routing and a full Clifford gate set, a sharp advance beyond storing protected logical memory. The demonstration represents a vital progression in superconducting surface-code experiments, paving the way for more complex quantum algorithms and fault-tolerant computation. Reusable qubit operations enable flexible surface code manipulation A new breakthrough hinged on developing a ‘primitive layer’ of reusable operations for manipulating surface-code patches; this is akin to a mosaic artist mastering a few key tile arrangements that can then be combined to create complex designs. Surface codes are a leading approach to quantum error correction, encoding logical qubits using multiple physical qubits arranged in a two-dimensional lattice. Protecting quantum information requires maintaining the delicate superposition and entanglement of qubits, which are highly susceptible to environmental noise. Surface codes achieve this by distributing the quantum information across the lattice and encoding it in the correlations between qubits. The developed primitive layer allows for dynamic rearrangement of these encoded qubits without destroying the encoded information. Merge, split, expansion, shrinkage, and deformations mediated by domain walls and twist defects allowed for precise reshaping of sections of the qubit grid without disrupting the encoded quantum information. Domain walls represent boundaries between regions with different logical properties, while twist defects introduce controlled changes in the lattice structure. These operations were not one-off adjustments but repeatable building blocks, enabling the creation of more complex logical operations. The ability to dynamically reconfigure the surface code is crucial for implementing complex quantum algorithms, as it allows for efficient routing of quantum information and allocation of resources. A 107-qubit superconducting quantum processor successfully demonstrated key elements of patch-based surface-code logical processing. Combining these reusable operations created logical state routing, a controlled-NOT gate, and single-qubit Hadamard and phase gates, forming a complete set for Clifford computations. Clifford computations are a subset of quantum computations that can be efficiently simulated on classical computers, but they serve as a crucial benchmark for evaluating the performance of quantum error correction schemes. All operations utilised distance-three rotated surface-code patches with multi-round syndrome extraction and neural-network decoding, avoiding post-selection of data. The system’s performance showcases the potential for actively processing quantum information, rather than merely storing it. Syndrome extraction is the process of measuring the errors that have occurred in the qubits without collapsing the quantum state. Multi-round syndrome extraction improves the accuracy of error detection, while neural-network decoding provides a sophisticated method for inferring the most likely error configuration based on the measured syndromes. Avoiding post-selection, where only successful runs are considered, is vital for obtaining realistic performance metrics. High fidelity logical gate operation via reusable surface code patch manipulation A logical controlled-NOT gate fidelity of 92.7% was achieved on a 107-qubit processor, representing a substantial improvement over previous demonstrations which primarily focused on maintaining stable, protected logical memory.

The team from Hefei National Research Centre for Physical Sciences at the Microscale and collaborating institutions moved beyond storing quantum information to actively processing it using a patch-based approach. This involved manipulating sections of a qubit grid, known as surface-code patches, without disrupting the encoded quantum data. The fidelity of a quantum gate measures the accuracy with which it performs its intended operation. A higher fidelity indicates a lower error rate and a more reliable quantum computation. Achieving high-fidelity logical gates is essential for building fault-tolerant quantum computers, as it allows for the accumulation of many operations without overwhelming the error correction capabilities. This layer incorporates merge and split operations, allowing patches of qubits to be joined or separated to infer logical parities. Patch expansion and shrinkage alter the size of qubit groupings, while deformations mediated by domain walls and twist defects enable precise reshaping of the qubit layout. Logical state routing was successfully implemented, alongside the logical controlled-NOT, Hadamard, and phase gates, forming a complete Clifford-generating set of operations essential for quantum computation. Logical state routing is the process of moving a logical qubit from one location on the lattice to another, which is necessary for performing complex quantum algorithms. These operations were performed on distance-three rotated surface-code patches, utilising multi-round syndrome extraction and neural-network decoding without needing to discard any data during processing. Rotating the surface code can improve its performance by reducing the impact of certain types of errors. The use of neural-network decoding represents a significant advance in error correction, as it allows for more accurate and efficient decoding of the error syndromes. Logical qubit operations demonstrate initial fault-tolerant computation with limited error The demonstration of active logical operations marks a key step beyond safeguarding quantum information; it allows for genuine computation within a protected environment. However, the current architecture relies on distance-three surface codes, a relatively basic level of error correction, and the researchers acknowledge that scaling to higher distances presents significant challenges. Distance refers to the size of the surface code lattice, with larger distances providing greater error correction capabilities. While these codes offer a starting point, achieving truly strong fault tolerance demands substantially more qubits and increasingly complex control systems to manage the growing computational overhead. The computational overhead arises from the need to encode each logical qubit using multiple physical qubits and to perform the necessary error correction operations. It is important to acknowledge that scaling these initial surface codes remains a formidable task. Maintaining coherence and controlling errors become increasingly difficult as the number of qubits increases. This moves superconducting qubit technology closer to practical quantum computing, even with the current limitations in qubit numbers and control complexity. Further research will focus on addressing these scaling challenges and exploring more advanced error correction techniques, such as topological codes with higher distances and more efficient decoding algorithms. Active, patch-based fault-tolerant logical operations were demonstrated, moving beyond storing protected quantum information. Implementing this reusable approach enabled precise reshaping of the qubit layout and logical state routing, alongside a complete Clifford gate set, establishing a foundation for building more complex quantum processors capable of performing calculations with active error correction, and raising the question of how to improve durability through increased code distances. The ability to perform logical operations with high fidelity is a crucial step towards realising the full potential of quantum computation, enabling the development of algorithms that can solve problems intractable for classical computers. The researchers successfully demonstrated active, patch-based fault-tolerant logical operations using 107 superconducting qubits. This achievement represents progress beyond simply storing protected quantum information, as logical state routing and a complete set of Clifford gates were implemented. These operations were performed on distance-three surface-code patches with multi-round syndrome extraction and neural-network decoding, without relying on post-selection. The authors intend to address scaling challenges and explore more advanced error correction techniques to improve code distances and overall system performance. 👉 More information 🗞 Surface code logical operations on a superconducting quantum processor 🧠 ArXiv: https://arxiv.org/abs/2607.01473 Stay current. See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals. Tags: