pith. sign in

arxiv: 2407.16176 · v2 · submitted 2024-07-23 · 🪐 quant-ph

Quantum memory based on concatenating surface codes and quantum Hamming codes

Menglong Fang , Daiqin Su This is my paper

Pith reviewed 2026-05-23 22:36 UTC · model grok-4.3

classification 🪐 quant-ph
keywords quantum error correctionsurface codequantum Hamming codeconcatenated codesfault toleranceerror thresholdquantum memorylogical error rate
0
0 comments X

The pith

Concatenating surface codes with quantum Hamming codes suppresses logical errors more effectively than surface codes at comparable resource levels.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The authors investigate a quantum memory constructed by concatenating surface codes with quantum Hamming codes. They find that this hybrid code maintains a high error threshold close to that of the surface code while achieving lower logical error rates than pure surface codes when the physical qubit overhead is similar. The improvement becomes noticeable at intermediate sizes of the quantum memory. This setup is evaluated for threshold, overhead, and decoding time, positioning it as a practical step toward fault-tolerant quantum computation.

Core claim

The concatenation of surface codes and quantum Hamming codes as a quantum memory achieves a high error threshold that can be pushed up to the surface code threshold. Under comparable resource overhead, the concatenated codes suppress logical errors to a much lower level than the surface codes, with the advantage appearing at intermediate scales.

What carries the argument

The specific concatenation scheme that layers quantum Hamming codes over surface code patches to combine their error-correcting properties.

If this is right

  • The error threshold remains high and can approach the surface code threshold.
  • Logical error suppression exceeds that of surface codes at similar qubit counts.
  • The benefit emerges for intermediate-scale quantum memories.
  • It enables small-scale fault-tolerant circuits with reduced resources.
  • Decoding efficiency is preserved due to the code structures.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • This could allow near-term quantum hardware to reach useful logical error rates with fewer physical qubits.
  • Other code combinations might be explored to further optimize the threshold-overhead tradeoff.
  • If implemented, it would bridge the gap between current noisy devices and large-scale fault tolerance.
  • The decoding time estimates suggest feasibility for real-time error correction in algorithms.

Load-bearing premise

The specific way of concatenating the two codes preserves the surface code's error threshold without new dominant failure modes or excessive decoding complexity.

What would settle it

Numerical Monte Carlo simulations of the concatenated code's logical error rate versus physical error rate for memory sizes around 100-1000 qubits, compared directly to equivalent surface code sizes, to check if the logical error is lower.

Figures

Figures reproduced from arXiv: 2407.16176 by Daiqin Su, Menglong Fang.

Figure 1
Figure 1. Figure 1: FIG. 1: Error correction circuit for a block of the [[7 [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2: Correlation coefficients for pairs of logical qubits for [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3: Logical error rate of concatenated quantum Hamming [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4: Error thresholds for surface-Hamming codes. The code distance [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5: Comparison of logical error rates between surface-Hamming codes and surface codes without concatenation, given a [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6: Comparison of decoding time for concatenated quan [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7: Estimated error probability of different logical qubits [PITH_FULL_IMAGE:figures/full_fig_p008_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8: Error probability for different number of failure [PITH_FULL_IMAGE:figures/full_fig_p009_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: FIG. 9: Correlation coefficients for a set of logical qubits and individual logical qubits. These sets are (a) 2-qubit sets, (b) [PITH_FULL_IMAGE:figures/full_fig_p010_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: FIG. 10: Logical error rate for concatenated quantum Ham [PITH_FULL_IMAGE:figures/full_fig_p012_10.png] view at source ↗
read the original abstract

Designing quantum error correcting codes that promise a high error threshold, low resource overhead and efficient decoding algorithms is crucial to achieve large-scale fault-tolerant quantum computation. The concatenated quantum Hamming code is one of the potential candidates that allows for constant space overhead and efficient decoding. We study the concatenation of surface codes with quantum Hamming codes as a quantum memory, and estimate its error threshold, resource overhead and decoding time. A high error threshold is achieved, which can in principle be pushed up to the threshold of the surface code. Furthermore, the concatenated codes can suppress logical errors to a much lower level than the surface codes, under the assumption of comparable amount of resource overhead. The advantage in suppressing errors starts to show for a quantum memory of intermediate scale. Concatenating surface codes with quantum Hamming codes therefore provides a promising avenue to demonstrate small-scale fault-tolerant quantum circuits in the near future, and also paves a way for large-scale fault-tolerant quantum computation.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript studies the concatenation of surface codes with quantum Hamming codes as a quantum memory. It estimates the error threshold, resource overhead, and decoding time of the resulting code, claiming a high threshold that can in principle reach the surface-code value, substantially lower logical error rates than standalone surface codes at comparable overhead, and that the error-suppression advantage appears already at intermediate scales.

Significance. If the threshold preservation and overhead claims are borne out by the underlying calculations, the construction would supply a concrete route to improved logical-error suppression without a prohibitive increase in physical resources, potentially useful for near-term fault-tolerant demonstrations and as a stepping stone toward larger-scale fault tolerance.

major comments (2)
  1. [threshold estimation section] The central claim that the concatenated threshold can be pushed up to the surface-code threshold (abstract and threshold-estimation section) rests on the unshown assertion that the outer Hamming layer does not introduce inter-layer correlated errors or decoder incompatibilities that would lower the effective threshold; an explicit check that the surface-code stabilizers remain intact after the concatenation map and that the joint decoder fails only at rates below the inner-code protection is required.
  2. [resource-overhead section] The statement that logical-error suppression is superior “under the assumption of comparable amount of resource overhead” (abstract and resource-overhead section) is load-bearing for the intermediate-scale advantage claim; a concrete qubit-count or gate-count comparison at fixed logical-qubit number (e.g., for distance-5 or distance-7 surface codes plus Hamming layer) must be supplied to substantiate that the overhead remains comparable rather than growing faster than the surface-code baseline.
minor comments (2)
  1. [decoding-time subsection] The decoding-time estimate should specify whether the reported figure includes the cost of the outer Hamming decoder or only the inner surface-code decoder; the distinction affects the practicality claim.
  2. [figures] Figure captions for the threshold and overhead plots should state the precise error model (depolarizing, circuit-level, etc.) and the number of Monte-Carlo shots used, to allow direct comparison with existing surface-code literature.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive suggestions. The two major comments identify points where additional explicit verification would strengthen the presentation; we address each below and will revise the manuscript accordingly.

read point-by-point responses

  1. Referee: [threshold estimation section] The central claim that the concatenated threshold can be pushed up to the surface-code threshold (abstract and threshold-estimation section) rests on the unshown assertion that the outer Hamming layer does not introduce inter-layer correlated errors or decoder incompatibilities that would lower the effective threshold; an explicit check that the surface-code stabilizers remain intact after the concatenation map and that the joint decoder fails only at rates below the inner-code protection is required.

    Authors: We agree that an explicit verification of stabilizer preservation and decoder compatibility is needed to support the threshold claim. In the revised manuscript we will add a short subsection that (i) shows the surface-code stabilizers are unchanged by the concatenation map and (ii) derives the condition under which the joint decoder’s logical failure rate is governed solely by the inner surface-code protection, thereby confirming that the outer Hamming layer does not introduce a lower threshold. revision: yes

  2. Referee: [resource-overhead section] The statement that logical-error suppression is superior “under the assumption of comparable amount of resource overhead” (abstract and resource-overhead section) is load-bearing for the intermediate-scale advantage claim; a concrete qubit-count or gate-count comparison at fixed logical-qubit number (e.g., for distance-5 or distance-7 surface codes plus Hamming layer) must be supplied to substantiate that the overhead remains comparable rather than growing faster than the surface-code baseline.

    Authors: We concur that a direct, quantitative comparison is required. The revised resource-overhead section will contain a table (and accompanying text) that reports total physical-qubit counts for a fixed number of logical qubits, comparing the concatenated construction (distance-5 surface code plus Hamming layer) against standalone surface codes of distances chosen to achieve comparable logical error rates. This will make the “comparable overhead” assumption concrete and allow readers to assess the claimed advantage. revision: yes

Circularity Check

0 steps flagged

No circularity; derivation relies on standard code concatenation and threshold estimation without self-referential reduction

full rationale

The abstract and available text describe a study of concatenating surface codes with quantum Hamming codes, estimating threshold, overhead, and decoding time via standard methods. No equations, fitting procedures, or self-citations are shown that reduce a claimed prediction or threshold result to an input by construction. The central claims (high threshold approachable to surface-code value, lower logical errors at comparable overhead) are presented as outcomes of the concatenation scheme rather than definitions or renamings of inputs. No load-bearing uniqueness theorems or ansatzes imported from prior author work are visible. This is the normal case of a self-contained numerical/simulation study whose results stand or fall on external verification rather than internal redefinition.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review yields no visible free parameters, axioms, or invented entities; all such details are absent from the provided text.

pith-pipeline@v0.9.0 · 5689 in / 1030 out tokens · 17293 ms · 2026-05-23T22:36:02.811621+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

41 extracted references · 41 canonical work pages · 1 internal anchor

  1. [1]

    Denote the syn- dromes of the n-th block at level l as S(n) l = {si | i ∈ [0, kl − 1]}, where kl = l + 3 is the number of stabi- lizer generators

    Measure the syndromes at level l. Denote the syn- dromes of the n-th block at level l as S(n) l = {si | i ∈ [0, kl − 1]}, where kl = l + 3 is the number of stabi- lizer generators. Here si ∈ {0, 1}, and si = 1 when 3 FIG. 1: Error correction circuit for a block of the [[7 , 1, 3]] quantum Hamming code in our numerical simulation. Errors occur only on seve...

    work page

  2. [2]

    Calculate N = kl−1X i=0 si2i, where N is the number that indicates which level- (l − 1) logical qubit has suffered an error

    work page

  3. [3]

    Execute step 2 in parallel on all blocks in level l and correct the error by applying the operator ¯X l−1 N,n to each block

    work page

  4. [4]

    The decoding algorithm runs on each block in O(kl) time, implying that decoding at each level also requires O(kl) time with parallel execution

    If level l is not the final level, repeat the step 1 to 3 at the next level. The decoding algorithm runs on each block in O(kl) time, implying that decoding at each level also requires O(kl) time with parallel execution. When this algorithm operates on a concatenated quantum Hamming code with a total of ℓ levels, the overall time complexity is O(ℓ2). C. L...

    work page

  5. [5]

    set of processed operators

    If the generator g1 commutes with all other genera- tors, set it aside in a “set of processed operators”

    work page

  6. [6]

    set of processed operators

    If the generator g1 anticommutes with another gener- ator gj, modify the remaining generators as follows: ∀i ∈ {2, · · · , m}, i ̸= j gi →    gi, [ gi, g1 ] = 0, [ gi, gj ] = 0; gi, {gi, g1} = 0, {gi, gj} = 0; gig1, [ gi, g1 ] = 0, {gi, gj} = 0; gigj, {gi, g1} = 0, [ gi, gj ] = 0. After the modification, all other generators are com- mute with both g...

    work page

  7. [7]

    Execute the above procedure recursively to the re- maining generators. By applying this algorithm to the generator matrix G, where each row represents an independent generator of the codewords, we obtain a matrix where each row represents either a logical operator or a stabilizer. After deleting the rows representing the stabilizers, we obtain a matrix de...

    work page

  8. [8]

    The error bars are calculated at the 95% confidence interval

    The threshold is about pth ≈ 1.5%. The error bars are calculated at the 95% confidence interval. a higher threshold can be achieved by concatenating a surface code of a larger size with quantum Hamming codes. The decoding algorithms for the surface codes and quantum Hamming codes are different, therefore a com- plete decoding process is divided into two s...

    work page

  9. [9]

    We assume that this also holds when |ES| > 1. In particular, we assume that the probability of errors with equal cardinal number is approximately identical, namely, P (ES) = pi (B2) for all ES satisfying |ES| = i, where i ∈ [1, K] with K the number of logical qubits. Therefore, under this as- sumption, we only need to verify whether pi ≤ pi for a given p....

    work page

  10. [10]

    P. W. Shor, in Proceedings of 37th conference on founda- tions of computer science (IEEE, 1996), pp. 56–65

    work page 1996

  11. [11]

    Aharonov and M

    D. Aharonov and M. Ben-Or, in Proceedings of the twenty- 12 FIG. 10: Logical error rate for concatenated quantum Ham- ming codes. The solid lines represent the logical error rate derived from assuming independent errors, while the dotted lines represent the logical error rate derived from a full decod- ing processing that takes into account correlated err...

    work page 1997

  12. [12]

    Knill, R

    E. Knill, R. Laflamme, and W. H. Zurek, Science 279, 342 (1998)

    work page 1998

  13. [13]

    A. Y. Kitaev, Annals of physics 303, 2 (2003)

    work page 2003

  14. [14]

    A. G. Fowler, M. Mariantoni, J. M. Martinis, and A. N. Cleland, Phys. Rev. A 86, 032324 (2012), URL https: //link.aps.org/doi/10.1103/PhysRevA.86.032324

    work page doi:10.1103/physreva.86.032324 2012

  15. [15]

    N. P. Breuckmann and J. N. Eberhardt, PRX Quantum 2, 040101 (2021), URL https://link.aps.org/doi/10. 1103/PRXQuantum.2.040101

    work page 2021

  16. [16]

    Gottesman, Quantum Information & Computation 14, 1338 (2014)

    D. Gottesman, Quantum Information & Computation 14, 1338 (2014)

    work page 2014

  17. [17]

    N. P. Breuckmann and B. M. Terhal, IEEE Transactions on Information Theory 62, 3731 (2016)

    work page 2016

  18. [18]

    Higgott and N

    O. Higgott and N. P. Breuckmann, arXiv preprint arXiv:2308.03750 (2023)

    work page arXiv 2023

  19. [19]

    Tillich and G

    J.-P. Tillich and G. Z´ emor, IEEE Transactions on Infor- mation Theory 60, 1193 (2013)

    work page 2013

  20. [20]

    N. P. Breuckmann and J. N. Eberhardt, IEEE Transac- tions on Information Theory 67, 6653 (2021)

    work page 2021

  21. [21]

    A. A. Kovalev and L. P. Pryadko, Phys. Rev. A 88, 012311 (2013), URL https://link.aps.org/doi/10. 1103/PhysRevA.88.012311

    work page 2013

  22. [22]

    Panteleev and G

    P. Panteleev and G. Kalachev, Quantum 5, 585 (2021)

    work page 2021

  23. [23]

    Lin and L

    H.-K. Lin and L. P. Pryadko, Phys. Rev. A 109, 022407 (2024), URL https://link.aps.org/doi/10. 1103/PhysRevA.109.022407

    work page 2024

  24. [24]

    Wang, H.-K

    R. Wang, H.-K. Lin, and L. P. Pryadko, in 2023 12th International Symposium on Topics in Coding (ISTC) (IEEE, 2023), pp. 1–5

    work page 2023

  25. [25]

    Panteleev and G

    P. Panteleev and G. Kalachev, in Proceedings of the 54th Annual ACM SIGACT Symposium on Theory of Comput- ing (2022), pp. 375–388

    work page 2022

  26. [26]

    Leverrier and G

    A. Leverrier and G. Z´ emor, in2022 IEEE 63rd Annual Symposium on Foundations of Computer Science (FOCS) (IEEE, 2022), pp. 872–883

    work page 2022

  27. [27]

    M. A. Tremblay, N. Delfosse, and M. E. Beverland, Phys. Rev. Lett. 129, 050504 (2022), URL https://link.aps. org/doi/10.1103/PhysRevLett.129.050504

    work page doi:10.1103/physrevlett.129.050504 2022

  28. [28]

    Bravyi, A

    S. Bravyi, A. W. Cross, J. M. Gambetta, D. Maslov, P. Rall, and T. J. Yoder, Nature 627, 778 (2024)

    work page 2024

  29. [29]

    Q. Xu, J. P. Bonilla Ataides, C. A. Pattison, N. Raveen- dran, D. Bluvstein, J. Wurtz, B. Vasi´ c, M. D. Lukin, L. Jiang, and H. Zhou, Nature Physics pp. 1–7 (2024)

    work page 2024

  30. [30]

    Berthusen, D

    N. Berthusen, D. Devulapalli, E. Schoute, A. M. Childs, M. J. Gullans, A. V. Gorshkov, and D. Gottesman, arXiv preprint arXiv:2404.17676 (2024)

    work page arXiv 2024

  31. [31]

    Yamasaki and M

    H. Yamasaki and M. Koashi, Nature Physics 20, 247–253 (2024), ISSN 1745-2481, URL http://dx.doi.org/10. 1038/s41567-023-02325-8

    work page 2024

  32. [32]

    Yoshida, S

    S. Yoshida, S. Tamiya, and H. Yamasaki, arXiv preprint arXiv:2402.09606 (2024)

    work page arXiv 2024

  33. [33]

    A. M. Steane, Phys. Rev. A 54, 4741 (1996), URL https: //link.aps.org/doi/10.1103/PhysRevA.54.4741

    work page doi:10.1103/physreva.54.4741 1996

  34. [34]

    M. A. Nielsen and I. L. Chuang, Quantum computation and quantum information (Cambridge university press, 2010)

    work page 2010

  35. [35]

    Steane, Proceedings of the Royal Society of London

    A. Steane, Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences 452, 2551 (1996), https://royalsocietypublishing.org/doi/pdf/10.1098/rspa.1996.0136, URL https://royalsocietypublishing.org/doi/abs/ 10.1098/rspa.1996.0136

    work page doi:10.1098/rspa.1996.0136 1996

  36. [36]

    A. R. Calderbank and P. W. Shor, Phys. Rev. A 54, 1098 (1996), URL https://link.aps.org/doi/10.1103/ PhysRevA.54.1098

    work page 1996

  37. [37]

    M. M. Wilde, Phys. Rev. A 79, 062322 (2009), URL https://link.aps.org/doi/10.1103/PhysRevA.79. 062322

    work page doi:10.1103/physreva.79 2009

  38. [38]

    M. P. da Silva, C. Ryan-Anderson, J. M. Bello-Rivas, A. Chernoguzov, J. M. Dreiling, C. Foltz, F. Frachon, J. P. Gaebler, T. M. Gatterman, L. Grans-Samuelsson, et al., Demonstration of logical qubits and repeated error correction with better-than-physical error rates (2024), 2404.02280, URL https://arxiv.org/abs/2404.02280

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  39. [39]

    Acharya, L

    R. Acharya, L. Aghababaie-Beni, I. Aleiner, T. I. Ander- sen, M. Ansmann, F. Arute, K. Arya, A. Asfaw, N. As- trakhantsev, J. Atalaya, et al., Quantum error correction below the surface code threshold (2024), 2408.13687, URL https://arxiv.org/abs/2408.13687

    work page arXiv 2024

  40. [40]

    Higgott, Pymatching: A python package for decoding quantum codes with minimum-weight perfect matching (2021), 2105.13082, URL https://arxiv.org/abs/2105

    O. Higgott, Pymatching: A python package for decoding quantum codes with minimum-weight perfect matching (2021), 2105.13082, URL https://arxiv.org/abs/2105. 13082

    work page arXiv 2021

  41. [41]

    C. A. Pattison, A. Krishna, and J. Preskill, arXiv preprint arXiv:2303.04798 (2023)

    work page arXiv 2023