Modern approaches to quantum computing based on topological codes require a global decoder that processes the syndromes of all qubits. This creates a bottleneck: the classical computer must process data faster than the system grows. Nearby in the archive are studies of quantum error correction, quantum memory, and stochastic decoders, but all of them either do not support computation or require growing resources. This paper stands apart: it is the first to combine local decoding with a full set of fault-tolerant operations in two dimensions.
The authors construct a cellular automaton that moves defects (violated parity checks) toward the corner of their cluster until they annihilate. Messages from defects propagate in three quadrants, guiding the motion. To handle phenomenological noise, a stack of coupled decoder copies is built, in which defects advance up the stack according to timers that grow exponentially. For general stabilizer codes, the polynomial formalism of Haah and Gröbner bases are used, which makes it possible to parallelize polynomial division. A hierarchical version of the decoder reduces the resource density to a constant, at the cost of translation invariance.
The first fully local fault-tolerant quantum computer in two dimensions based on the surface code has been built. The decoder preserves logical information for a time that grows exponentially with the code distance, and it runs continuously during state injection, stabilizer preparation, lattice surgery, and transversal readout. It is proven that every translation-invariant topological stabilizer code is locally decodable under phenomenological noise. Numerically, the threshold for the surface code came out to about 1.1% under independent noise.
Unlike neighboring works, which focus on quantum memory or require polylogarithmic resources, this architecture achieves a constant density of classical resources and supports universal computation. This is the first demonstration that locality is not a fundamental obstacle to scalable quantum computing. The result also broadens the class of locally decodable quantum memories, including fracton codes, and offers a unified principle for constructing local decoders.
Possible directions include optimizing practical performance with neural cellular automata, removing the assumption that classical computation is error-free, and creating a fully stable architecture. Applying the approach to non-Abelian codes and local real-time decoding for schemes without state distillation are also promising.
The result will influence the design of quantum processors based on superconducting qubits and quantum dots, making it possible to perform error correction entirely on-chip at cryogenic temperatures.
The immediate steps include adapting the decoder to non-Abelian codes, developing versions that are robust to noise in classical computation, and numerically estimating thresholds for hierarchical constructions.
The work is directly connected to the open problem of classifying nonequilibrium quantum phases of matter and to the question of which quantum memories admit local decoding. The proof that all translation-invariant stabilizer codes are locally decodable is a significant step in this area.
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
The work builds a cellular-automaton decoder for the surface code that runs continuously during all operations, including state injection and lattice surgery. The decoder uses a constant number of classical bits per node and proves an error threshold of about 1.1%. It is also shown that any translation-invariant topological stabilizer code admits local decoding.
Where its strength is
The strength lies in the fact that locality is not a fundamental obstacle to scalable quantum computing: a concrete mechanism is proposed with a constant density of classical resources that supports universal operations.
Where it could go next
Worth trying
Why this may matter
If local decoders become practical, this will substantially reduce the engineering requirements for classical control electronics and bring scalable fault-tolerant quantum computers closer.
Nearby works in our archive
The closest work in the archive sits at 0.62 by meaning — the usual level for our corpus.
The decoder uses a principle first applied in the Bravyi-Haah broom algorithm: defects are swept toward the corner of a cluster, like trash to a wall.
The idea of local error correction without a central controller is reminiscent of the decentralized systems in Robert Ibatullin's novel "The Rose and the Worm," where computing networks self-repair after damage.