Concatenated Quantum Codes

Emanuel Knill; Raymond Laflamme

REVIEW

Concatenated Quantum Codes

Emanuel Knill, Raymond Laflamme

Quantum Physics · Submitted 8 Aug 1996

Pith's verdict: The concatenation argument holds up and the constants are conservative, not wrong; the real caveat is the noise model, which the authors state plainly.

T0 · AI review

How this review was made: An AI referee (selfhosted/deepseek-v4.1-flash) read the paper on November 13, 2026 and answered a fixed list of questions about it. T0 is the starting level every paper gets. The levels describe how deeply Pith has checked the paper, never the authors.

T0 AI review An AI read the paper and answered a fixed list of questions.
T1 Audit Automatic checks that the references and files the paper relies on exist and match.
T2 Compute The paper's code or calculations re-run with the same results.
T3 Formal statement The claims written out in a formal math language a computer can read.
T4 Proof Those claims proved, and the proof checked by computer.

How Pith checks papers →

· 4 major objections · 6 minor · Ask about this paper · The compiled record · Read the full review →

Challenge this review Comment (0) Read on arXiv ↗

Cite ▾

arXiv IDquant-ph/9608012 (version 1)

Pith ID4LQPGIAF, Pith's permanent record of this paper and its review

CitationEmanuel Knill, Raymond Laflamme. "Concatenated Quantum Codes." arXiv:quant-ph/9608012 (1996). https://pith.science/paper/4LQPGIAF

Cite this reviewBibTeX for Pith's review of this paper

Claim paper

Why Pith says this

Highlight any text to comment on it.

Editor's note

read the letter

This is the paper that turned 'you need arbitrarily good gates' into 'you need good enough gates.' Hierarchical concatenation with recovery at every level, plus an induction that yields a constant threshold, is the contribution, and it holds up. Shor's scheme [4] needed polylogarithmically small error; this needs a fixed c times epsilon, independent of storage time or distance. The overhead bookkeeping in Theorem 4.2 is clean and worth reading on its own: n^{log_r l} qubits, (r+1)/(r-1) n parallel operations, n^{1+log_r l} basic operations. The error-propagation lemmas in Section 2.2 are the machinery that makes the induction go, and they are stated and proved carefully rather than waved through. The five-qubit worked example in Section 4.4 is a genuine service; most threshold papers leave the constants implicit forever. On the stress-test note about the 20 eps_d^2 term: I checked it, and it is not a discrepancy. Theorem 4.3's second inequality is 2*C(l,e+1)*eps^{e+1}; for l=5, e=1 that is 2*10*eps^2 = 20 eps^2. The exact sum leads with 10 eps^2, so the factor of two is the stated geometric-series bound, valid for eps <= 1/2. Conservative, not an error. Using 10 would nudge the threshold up slightly and change nothing structurally. The soft spots are the ones the authors name. The Section 2.3 assumptions, local independence, sequential independence, no leakage, are load-bearing, and correlated or non-Markovian noise breaks the guarantee. The rigorous threshold of 1/21600 gate amplitude is, in their own words, out of reach of foreseeable technology; the dissipation heuristic gives 5e-4, which is the number people actually quote. Assumption 4.2, that the recovery implementation is accurate against arbitrary single-qubit errors, is a real constraint the paper assumes rather than constructs; the authors point to [4,11] for that, honestly. The 30-gate count for the five-qubit recovery comes from [11], a preprint, and the numerical threshold rides on it. The proof of Theorem 4.1 is compressed at exactly the level-to-level induction step, which is where a referee should push. None of this undermines the result. This is for anyone working on fault tolerance, quantum memory, or channel coding; it is the reference for concatenation-based thresholds. It deserves a serious referee, and I would cite it without hesitation.

Referee report

Significance

If correct, this is a foundational result: it converts the qualitative hope that quantum error correction can be iterated into a quantitative threshold statement with explicit constants, and it does so for memory and transmission rather than only for computation. The paper is commendably explicit about what it assumes (§2.3) and about how far the rigorous numbers are from engineering reality (§5), and it separates rigorous from heuristic estimates rather than blending them. The amplitude-based accounting in §2.2 and Theorem 4.3 is the right tool for worst-case coherent error and is set out carefully, with the propagation lemmas stated and proved. The overhead theorem is derived rather than asserted, and the five-qubit worked example gives checkable inequalities rather than a hand-waved threshold. These are real strengths and should be weighed as such.

Major objections (4)

[§4.4, Theorem 4.4(2) versus Eq. (3)] There is a numerical inconsistency between the theorem statement and the inequality it is derived from. Eq. (3) is ε_d ≥ 90ε_p + 60ε_d², which with ε_d = ε gives ε_p ≤ (ε − 60ε²)/90 = ε(1 − 60ε)/90, whereas Theorem 4.4(2) states ε(1 − 60ε²)/90. The two differ by a factor of order two at ε = 1/120 (1/21600 versus roughly 1/10800), and the stated form is the stronger and hence invalid one. Please reconcile the theorem statement with Eq. (3); the derivation supports ε(1 − 60ε)/90. This does not affect the existence claim, but it is a quoted constant in a theorem and should be corrected.
[§4.1, proof of Theorem 4.1] The induction step is compressed into three sentences ('By the error propagation properties, the errors ... can be added'), and for a paper whose central claim is this theorem the composition should be spelled out. Specifically, the proof should state that the l lower-level errors are simultaneously of the product form required by assumption (4.2) precisely because of the local independence assumption, that Lemma 2.2 supplies the additivity of amplitudes across the encoding, the r−1 channel-plus-recovery blocks and the final channel-plus-decoding block, and that Lemma 2.3 is what allows the accumulated lower-level error to be substituted for the channel error I in assumption (4.2). As written, a reader must reconstruct which lemma licenses which step, and the count (r+1)ε_c is asserted rather than derived.
[§4.4, Eq. (3), coefficient of ε_d²] The coefficient 20 in the term 20ε_d² is not derived in the text, and it is worth stating explicitly that it is exactly the geometric bound of Theorem 4.3 with l = 5 and e = 1, namely 2·C(5,2)·ε² = 20ε², valid under the stated hypothesis ε_d ≤ (e+2)/(2(l−e−1)) = 1/2, which is the 'ε_d ≤ 1/2' quoted in §4.4. Making this step explicit would forestall the natural misreading that the leading term should be C(5,2)ε² = 10ε²; the factor of two is the geometric-series bound, not an error, and the threshold numbers in Theorem 4.4 are consistent with it.
[Abstract (abstract:s2) and §1] The abstract states the threshold result without qualification: 'to store or transmit a qubit with maximum error ε requires gates with error at most cε and storage or channel elements with error at most ε, independent of how long we wish to store the state or how far we wish to transmit it.' The result is conditional on the no-leakage, local independence and sequential independence assumptions of §2.3, and on the requirement r+1 ≤ ε_d/ε_c. Since the abstract is where the central claim is stated, it should carry at least a short qualifier pointing to the assumptions, as §5 does. This is a framing issue rather than a technical one, but it is the load-bearing sentence of the paper.

Minor objections (6)

[§4.4] The numerical thresholds rest on the count of at most 30 primitive operations for the five-qubit recovery, cited from reference [11], which is listed as a preprint. It would help readers to give the threshold as an explicit function of that count N (the recovery error enters as Nε_p), so the numbers can be rescaled if the implementation count changes.
[§1 and §4.2] The introduction says the method requires O(n^α) additional qubits 'with α dependent on the actual operational accuracy', whereas Theorem 4.2 gives the exponent as log_r l, determined by code length and repetition factor. The dependence on accuracy is only indirect, through the constraint that fixes r; please make the two statements consistent.
[§2.2] The symbol E is used for the encoding operation, for the environment, and for the error amplitude E(A,B). The overload is manageable in context but makes several sentences in §2.2 and §4.1 harder to parse than necessary; a distinct symbol for one of the three would help.
[§3] There are typographical errors that should be corrected, including 'there his no hope' in the paragraph on repeated recovery and 'the smallerst r of interest' in §4.3.
[§5] The dissipated-error heuristic estimate of about 5×10⁻⁴ is quoted without saying how it is obtained from the amplitude analysis. A sentence stating which amplitudes are replaced by probabilities and which inequalities are then used would make the heuristic reproducible.
[§4.4] The sentence 'Ignoring the overhead, we can achieve final error 3ε_c over any distance provided that ε_c/ε_d ≤ 1/3' would read more clearly if it were tied explicitly to r = 2 and r+1 = 3, since the factor 3 is the whole content of the requirement.

How Pith checks papers → · Full audit of this paper →

The paper

The paper claims

A qubit can be stored or sent arbitrarily far with fixed per-step error, provided each gate's error is a small constant multiple of the target error.

The reading

The paper asks whether a quantum memory or channel can hold a state for arbitrarily long times, or send it arbitrarily far, without demanding that every gate be arbitrarily accurate. Earlier proposals needed gate errors shrinking toward zero as the storage time or distance grew. Knill and Laflamme argue that this requirement can be removed: if storage or channel elements have error amplitude at most $epsilon$, gates with error amplitude at most $c,epsilon$ suffice for any duration or distance, with overhead polynomial in the number of time steps. The construction is recursive, so the interesting question becomes how large the constant $c$ can be, and the paper answers that with both a general theorem and an explicit calculation for the five-qubit code.

What carries the argument

Concatenation with hierarchically implemented recovery. The code is applied, then the same procedure is applied inside each of its qubits before the outer recovery runs, so that a fault anywhere in a recovery circuit is itself caught by the level below. The proof machinery is the paper's error-amplitude formalism: a generalized unitary operator $A$ is compared to an intended $B$ by writing $A = lambda B' + E$ with $|E| le epsilon$, and three lemmas show that these amplitudes compose additively under sequential operations and do not increase when ancillas are discarded. Additivity is what makes the induction close: level $h+1$ behaves exactly like level $1$ with the waiting period replaced by a level-$h$ procedure whose error is already known to be at most $epsilon_d$. Theorem 4.3 supplies the other half, bounding the residual error after recovery by the tail $sum_{i>e} binom{l}{i} epsilon^i$ of the binomial expansion, which is small whenever the per-qubit error is small compared with the code's correction capacity.

Extended reading notes

Core claim

The central claim is that concatenated quantum coding turns a fixed, non-vanishing level of operational error into arbitrarily long protection. The recipe: take a short one-error-correcting code, encode a qubit into it, apply the next lower level of the same procedure to each of the resulting qubits, then run the code's recovery operation, and repeat this nesting to a depth $h$. Theorem 4.1 proves by induction on the level that if encoding, recovery and decoding each carry error amplitude at most $epsilon_c$, and the channel or waiting period carries at most $epsilon_d$, then the whole level-$h$ procedure differs from the identity by at most $(r+1)epsilon_c$, provided $r+1 le epsilon_d/epsilon_c$. Because the bound per level does not grow with the level, the accumulated error stays fixed while the number of protected time steps grows as $r^h$. Theorem 4.2 prices this in resources: $n$ waiting or transmission periods need $n^{log_r l}$ qubits, fewer than $frac{r+1}{r-1}n$ parallel operations, and fewer than $n^{1+log_r l}$ basic encoding, recovery and decoding operations, where $l$ is the code length. Theorem 4.4 instantiates the scheme with the five-qubit code: for target error $epsilon ge 1/120$, storage error $1/120$ and primitive gate error $1/21600$ suffice; for $epsilon < 1/120$, storage error $epsilon$ and primitive gate error $epsilon(1-60epsilon^2)/90$ suffice.

Load-bearing premise

The guarantee rests on the noise model of Section 2.3: each qubit must stay in its two-dimensional space, distinct qubits must evolve independently within a time step, and successive operations must fail independently, so that per-level error amplitudes simply add.

Editorial extensions

If this is right

Arbitrarily long quantum storage and arbitrarily long quantum transmission become engineering problems with a fixed accuracy target rather than problems requiring asymptotically perfect hardware.
Quantum repeaters spaced at regular intervals along a channel with enough parallel paths could, in principle, implement the scheme.
The threshold is stated as a concrete number: for the five-qubit code, gate error amplitude near $1/21600$ together with storage error near $1/120$ already supports arbitrary storage time at final error $1/120$.
Resource costs are polynomial, not exponential: qubit count scales as $n^{log_r l}$ and total basic operations as $n^{1+log_r l}$ in the number $n$ of protected periods.
The scheme works with any code implemented accurately enough; it does not require a code that corrects one specific error type perfectly.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The paper's own threshold numbers are pessimistic because its recovery-circuit error budget is a crude sum over primitive gates; a fault-tolerant compilation of the same recovery, as the authors suggest in the conclusion, would likely raise the tolerable gate error by orders of magnitude.
The induction's cleanliness depends on errors being uncorrelated across qubits and across time, so the natural next test is whether the same conclusion survives a bath with memory, where the per-level error could grow with the level rather than staying fixed.
Because the bound is on error amplitudes rather than probabilities, coherent cancellation is not assumed away; this makes the guarantee stronger than a probabilistic estimate but also means a physical implementation must control systematic over-rotations, not just random faults.
A practical corollary the paper leaves implicit: the depth $h$ needed for a given storage time grows only logarithmically, so the interesting engineering variable is the constant in the exponent, not the architecture.

Original abstract

One of the main problems for the future of practical quantum computing is to stabilize the computation against unwanted interactions with the environment and imperfections in the applied operations. Existing proposals for quantum memories and quantum channels require gates with asymptotically zero error to store or transmit an input quantum state for arbitrarily long times or distances with fixed error. In this report a method is given which has the property that to store or transmit a qubit with maximum error $\epsilon$ requires gates with error at most $c\epsilon$ and storage or channel elements with error at most $\epsilon$, independent of how long we wish to store the state or how far we wish to transmit it. The method relies on using concatenated quantum codes with hierarchically implemented recovery operations. The overhead of the method is polynomial in the time of storage or the distance of the transmission. Rigorous and heuristic lower bounds for the constant $c$ are given.

The compiled record

The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-09-23: 1 request, 4,949 input tokens.

What the abstract itself states, as probabilities: claims first 0.06 · states limitation 0.46 · human participants 0.02 · controlled comparison 0.05 · randomized assignment 0.01 · states quantitative result 0.81 · compares with prior results 0.76 · mentions code or data availability 0.02.

The abstract, sentence by sentence

arXiv listing abstract v1. Results and the main claim are underlined; show every role.

One of the main problems for the future of practical quantum computing is to stabilize the computation against unwanted interactions with the environment and imperfections in the applied operations. Existing proposals for quantum memories and quantum channels require gates with asymptotically zero error to store or transmit an input quantum state for arbitrarily long times or distances with fixed error. In this report a method is given which has the property that to store or transmit a qubit with maximum error $\epsilon$ requires gates with error at most $c\epsilon$ and storage or channel elements with error at most $\epsilon$, independent of how long we wish to store the state or how far we wish to transmit it. The method relies on using concatenated quantum codes with hierarchically implemented recovery operations. The overhead of the method is polynomial in the time of storage or the distance of the transmission. Rigorous and heuristic lower bounds for the constant $c$ are given.

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#	sentence	role	the authors' own work	states the main conclusion	strength
s0	One of the main problems for the future of practical quantum computing is to stabilize the computation against unwanted interactions with the environment and imperfections in the applied operations.	background 1.00	0.05	0.03	asserted 1.00
s1	Existing proposals for quantum memories and quantum channels require gates with asymptotically zero error to store or transmit an input quantum state for arbitrarily long times or distances with fixed error.	background 0.81	0.04	0.03	asserted 1.00
s2	In this report a method is given which has the property that to store or transmit a qubit with maximum error $\epsilon$ requires gates with error at most $c\epsilon$ and storage or channel elements with error at most $\epsilon$, independent of how long we wish to store the state or how far we wish to transmit it.	contribution 0.55	0.90	0.89	asserted 1.00
s3	The method relies on using concatenated quantum codes with hierarchically implemented recovery operations.	method 0.98	0.71	0.07	asserted 1.00
s4	The overhead of the method is polynomial in the time of storage or the distance of the transmission.	result 0.96	0.85	0.08	asserted 1.00
s5	Rigorous and heuristic lower bounds for the constant $c$ are given.	result 0.62	0.91	0.05	asserted 0.99

What kind of paper the classifier reports this is: method or tool 0.96 · evidence from proof or derivation 0.99.

What was examined, and what was not

sources read

arXiv listing abstract v1

not examined

full text
figures and tables
references and citing sentences
relations to other papers

levels run

P abstract

not run

S sections
T body sentences
C citations
R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

References

Forward citations

Cited by 26 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 65 citations worldwide. Full citation record

Phantom Codes: Hardness, Rate Optimal qLDPC Constructions, and Distance Limits
quant-ph 2026-09 accept novelty 8.0 of 10 partial

Phantom codes are rare and hard to recognize, but this paper builds them at the maximal logical qubit count with constant distance and proves a no-go bound for faster-growing logical dimensions.
Blocklet concatenation: Low-overhead fault-tolerant protocols for fusion-based quantum computation
quant-ph 2025-06 conditional novelty 8.0 of 10

Blocklet concatenation yields fusion-based quantum computing protocols with constant-sized resource states, erasure thresholds up to 19.1%, and footprint per logical qubit scaling better than surface codes.
A local automaton for the 2D toric code
quant-ph 2024-12 conditional novelty 8.0 of 10

A strictly local measurement-and-feedback decoder for the 2D toric code is built from hierarchical Tsirelson-type automata, preserving a logical qubit for times exponential in system size.
Fault-Tolerant Quantum Computation with Constant Overhead
quant-ph 2013-10 conditional novelty 8.0 of 10

Fault-tolerant quantum computation can, in principle, achieve constant qubit overhead asymptotically using quantum LDPC codes, conditional on a suitable code family existing.
Fault-tolerant quantum computation by anyons
quant-ph 1997-07 accept novelty 8.0 of 10

Anyons in a two-dimensional system with only local interactions give a quantum computer that is fault-tolerant by physics, realized through the toric code and nonabelian anyon models.

Show 21 more citing papers

Random Quantum LDPC Codes Approaching the Gilbert-Varshamov Bound
cs.IT 2026-10 accept novelty 7.0 of 10

Random quantum LDPC codes built by concatenating low-degeneracy AEL outer codes with random inner codes attain the quantum Gilbert-Varshamov distance bound and match random stabilizer codes on erasure and Pauli channels.
Learning to Decode Concatenated Quantum Codes with Hierarchical Message Passing
quant-ph 2026-08 conditional novelty 7.0 of 10

A neural message-passing decoder for concatenated quantum codes achieves substantially higher thresholds and lower failure rates than handcrafted decoders.
Merging-Based Quantum Repeater
quant-ph 2025-02 conditional novelty 7.0 of 10

A merging-based quantum repeater that patches failed operations by recycling entanglement achieves higher simulated secret key rates than swapping-based repeaters.
Quantum accuracy threshold for concatenated distance-3 codes
quant-ph 2005-04 accept novelty 7.0 of 10

Concatenating a distance-3 quantum code drives the logical failure rate doubly exponentially small whenever the per-location fault rate is below 2.73 x 10^-5, and that bound is now proven rather than asserted.
Fault-Tolerant Postselected Quantum Computation: Threshold Analysis
quant-ph 2004-04 conditional novelty 7.0 of 10

Concatenating four-qubit error-detecting codes with postselection tolerates gate errors above 1 percent, according to a heuristic computer-assisted threshold analysis.
Fault-Tolerant Postselected Quantum Computation: Schemes
quant-ph 2004-02 conditional novelty 7.0 of 10

A four-qubit error-detecting code with a spectator qubit, combined with transversal gates, purification, and concatenation, is proposed as a way to run postselected quantum computation with noisy gates.
Stitched surface codes: planar concatenation through GHZ corridors
quant-ph 2026-10 accept novelty 6.0 of 10

The stitched surface code uses GHZ corridors to measure outer-code stabilizers across four surface-code patches, yielding two logical qubits at distance 2t in a planar nearest-neighbor layout with hardware demonstration.
Satisfying Quantum Codes: Physics-Informed and Hardware-Aware Code Design with SAT Solvers
quant-ph 2026-08 conditional novelty 6.0 of 10

A SAT-based framework designs quantum codes from scratch or from physics symmetries, proves code design NP-complete, and finds distance-three surface-like codes with lower simulated logical error rates than the XZZX code.
Oraqle: An Empirical Analysis of Qubit Readout and Discriminators in Quantum Error Correction
quant-ph 2026-08 conditional novelty 6.0 of 10

Using real 5-qubit traces, this study shows readout windows can be cut to ~600 ns with negligible QEC penalty and small discriminators match large ones.
Arbitrary-Distance Quantum Error Correction with Gauss's Law for $\mathbb Z_2$ Lattice Gauge Theory
hep-lat 2026-07 accept novelty 6.0 of 10

Gauss's law constraints in Z2 lattice gauge theory can be made into quantum error-correcting codes of arbitrary distance, with provably optimal encoding rate within the constructed family.
Synthesis and Optimization of Encoding Circuits for Fault-Tolerant Quantum Computation
quant-ph 2026-05 conditional novelty 6.0 of 10

New search algorithms over stabilizer tableaus and modular assembly techniques yield encoders with up to 43% fewer two-qubit gates and 70% lower depth than prior constructions on tested stabilizer codes including qLDP...
Coupled-Layer Construction of Quantum Product Codes
quant-ph 2026-03 unverdicted novelty 6.0 of 10

Tensor and balanced product codes arise from a coupled-layer construction via anyon condensation on stacked constituent codes.
Growing Sparse Quantum Codes from a Seed
quant-ph 2025-07 conditional novelty 6.0 of 10

Conjoining only bit-flip and phase-flip repetition codes can generate any CSS code, and an iterative algorithm grows sparse subsystem codes with kd^2=O(n) worst-case scaling.
Enhancing Decoding Performance using Efficient Error Learning
quant-ph 2025-07 conditional novelty 6.0 of 10

Using about 1% of Pauli error rates from Cycle Error Reconstruction plus a heuristic completion, maximum-likelihood decoding of concatenated Steane codes achieves roughly 5 to 10 times lower logical error rates than f...
Quantum Error Correction Exploiting Degeneracy to Approach the Hashing Bound
quant-ph 2025-06 conditional novelty 6.0 of 10

A proposed quantum LDPC code with degeneracy-aware post-processing reaches FER 10^-4 at 9.45% depolarizing noise, approaching the hashing bound.
Learning Encodings by Maximizing State Distinguishability: Variational Quantum Error Correction
quant-ph 2025-06 unverdicted novelty 6.0 of 10

VarQEC uses a distinguishability loss as a machine-learning objective to variationally discover resource-efficient encoding circuits optimized for given noise models.
Logically Augmented Quantum LDPC Codes: Construction and Hierarchical Decoding
cs.IT 2026-11 conditional novelty 5.0 of 10

A logically augmented QLDPC construction that expurgates logical qubits via an auxiliary protection code, combined with a hierarchical list decoder that reduces logical error rate by up to 81% at modest rate cost.
Entanglement-Assisted Concatenated Quantum Codes: Parameters and Asymptotic Performance
quant-ph 2025-01 reject novelty 5.0 of 10

New families of concatenated entanglement-assisted quantum codes are claimed to beat known code tables, and the paper attempts to show such codes asymptotically reach the entanglement-assisted Gilbert-Varshamov bound.
Quantum repeaters: From quantum networks to the quantum internet
quant-ph 2022-12 conditional novelty 3.0 of 10

A survey that organizes quantum repeater architectures into three generations defined by whether loss and operation errors are suppressed probabilistically or deterministically, and reviews the experimental progress t...
Fault-tolerant quantum computation
quant-ph 1997-12 accept novelty 3.0 of 10

Fault-tolerant quantum computation is possible in principle if the error rate per gate and per storage step is below a critical accuracy threshold.
Design Automation in Quantum Error Correction
quant-ph 2025-07 conditional novelty 2.0 of 10

A comprehensive review of automated tools and methods for designing quantum error-corrected circuits, with case studies on T-gate optimization, surface-code layout, ML decoders, and verification.

How later papers used it

59 later papers cite it

Evidence: substantial. 58 of the citing papers share no author with it, and 7 of those carry a confident label from Pith's citation classifier.

Pith read 19 of the 93 citing sentences, choosing only those whose citation markers point at this paper alone, and wrote the paraphrases from them. The summary, the counts, the evidence level, the links and the quotes come from code; the summary counts the papers listed below. A later paper is listed under a label when Pith's citation classifier gives one of its sentences that label at probability 0.8 or more (0.95 for support), Pith's reading finds that the citing paper itself did it, and a second check confirms the paraphrase against the sentence. This section lists uses only. It does not list criticism or disagreement, so their absence here says nothing about how the paper has been received. Papers that share an author with this one are listed apart, under "By the same authors". A quote is the citing authors' own sentence, shown when their paper's licence is CC BY, CC BY-SA or CC0. JSON

1 later paper by other authors is listed below: 1 used its method or tool.

Uses its method or tool 1 of 4 papers with this label

Restricted its consideration to concatenations of this paper's kind, specifically of the 7-qubit Steane code with elementary gadgets constructed in another work.

Fault-tolerant Coding for Quantum Communication (2020) · 0.86 · paraphrased: the licence does not allow quoting

How the literature uses it

59 arXiv papers cite this in 92 sentences

Each citing sentence is taken from the citing paper's LaTeX source and typed by Pith's classifier. A label appears at probability 0.8 or more (0.95 for support), a contest at 0.7 or more; the quoted words are the citing authors'. How each label was checked · JSON

Papers citing it · 45 sentences without a confident label

Builds on it 1 paper

It provides a natural means for extending key concepts such as the fidelity of a quantum channel [ref] or the entanglement fidelity [ref] to the case of subsystem codes.

Topics in quantum information and the theory of open quantum systems · 7.5 Summary and outlook · 0.96

Uses its method or tool 3 papers

In the following we will restrict to the concatenations [ref] of the $7$-qubit Steane code for which elementary gadgets have been constructed in [ref] .

Fault-tolerant Coding for Quantum Communication · Analyzing noisy quantum circuits · 0.86
The 7-qubit Steane code introduced in [ref] is an error correcting code that can correct all single-qubit errors, and that can be concatenated to improve protection against errors [ref] .

Quantum communication and fault-tolerance · Fault-tolerance for quantum circuits · 0.94
We can then follow the standard procedure for finding the codewords and parameters of a concatenated code as in, e.g., Refs. [ref] : Replacing each $_in,C$-dimensional subsystem of the outer code by $_in$, the concatenated code hence has $_+ = (2^n_in)^q$ and $_+,C = 2^k_out$.

Existence of Pauli-like stabilizers for every quantum error-correcting code · Code parameters and codewords of a concatenated binary code · 0.80

Compares with it 2 papers

Comparison of approaches The main advantage of our scheme over existing ones [ref] is the superior scaling of the local resources with the distance, which is reduced from polynomial [ref] or polylogarithmic [ref] to constant.

Long-range big quantum-data transmission · Introduction · 1.00
Compared with other quantum concatenation schemes, such as concatenated quantum codes (CQC) [ref] , the component code selections of QTPCs are more flexible than those of CQCs.

On Quantum Tensor Product Codes · Introduction · 0.94

Cites it as background 34 papers

The idea of concatenated codes, originally described by Forney in a seminal book in 1966 [ref] , was introduced to quantum computation community three decades later [ref] .

Generalized Concatenated Quantum Codes · Introduction · 0.93
The concatenated Steane codes [ref] are constructed by recursively encoding each physical qubit of a Steane code into a logical qubit of the Steane code.

Polylog-time- and constant-space-overhead fault-tolerant quantum computation with quantum low-density parity-check codes · Concatenated Steane codes · 0.86
Similarly in QECCs, the concatenated quantum codes (CQCs), introduced by Knill and Laflamme in 1996 [ref] , are also effective for constructing good quantum codes.

Entanglement-assisted concatenated quantum codes · Introduction · 0.83

Discussion

Disagree with the review, or know something it missed? Leave a comment, highlight any passage above to comment on it, or challenge this review formally.