pith. sign in

arxiv: 2605.23236 · v1 · pith:HX6XTK7Mnew · submitted 2026-05-22 · 💻 cs.IT · math.IT

A Posterior MWPM Decoding Boosts the XYZ Planar Code

Zhiwei Wang , Liqi Wang This is my paper

Pith reviewed 2026-05-25 03:13 UTC · model grok-4.3

classification 💻 cs.IT math.IT
keywords surface codesquantum error correctionMWPM decoderbiased noiseXYZ planar codeposterior decodingthresholdslogical error rates
0
0 comments X

The pith

The XYZ planar code with posterior MWPM decoding achieves higher and more stable thresholds under biased noise than the standard surface code.

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

The paper introduces the XYZ planar code, a modification of the surface code, and extends the minimum-weight perfect matching decoder to a posterior version (pMWPM) that adds almost no complexity. It establishes that this combination produces higher thresholds and lower logical error rates across a wide range of bias parameters in an independent biased Pauli noise model. A reader would care because real quantum hardware often exhibits biased noise, so improved thresholds translate directly to fewer physical qubits needed for reliable logical operations. The gains are quantified explicitly, including a 36 percent threshold increase in the infinite-bias limit and stable performance at other bias values such as 15.5 percent at η=1.

Core claim

The XYZ planar code exhibits higher and more stable thresholds than the planar code under almost all bias conditions, while also achieving significantly lower logical error rates. Specifically, in the infinite-bias case, the threshold of the XYZ planar code is improved by about 36% compared to that of the surface code, and it maintains comparable or higher thresholds under other biases -- for example, the threshold reaches approximately 15.5% at bias η = 1 and 14.2% at η = 100.

What carries the argument

The XYZ planar code (a modified surface code) paired with the posterior MWPM (pMWPM) decoder that incorporates posterior probability information from the bias parameter η.

If this is right

  • pMWPM adapts to a wide range of modified surface codes with negligible extra decoding cost.
  • The XYZ planar code delivers lower logical error rates than the planar code at the same physical error rate under the tested biases.
  • Thresholds remain comparable or higher than the surface code at finite biases including η=1 and η=100.
  • The method indicates potential for codes in which Y operators act on larger numbers of data qubits.

Where Pith is reading between the lines

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

  • If the noise model holds in hardware, the reduced overhead could allow smaller-scale demonstrations of fault tolerance.
  • The same posterior-matching idea may transfer to other stabilizer codes that are not surface codes.
  • Varying the bias parameter during decoding could be tested as a way to handle time-varying noise in real devices.

Load-bearing premise

The physical noise consists of independent biased Pauli errors whose probabilities are fully captured by the single parameter η, and the pMWPM extension correctly folds the resulting posterior information into the matching without added complexity.

What would settle it

Monte Carlo simulations under infinite bias (η → ∞) that show the XYZ planar code threshold is not at least 36 percent higher than the surface-code threshold when decoded with pMWPM.

Figures

Figures reproduced from arXiv: 2605.23236 by Liqi Wang, Zhiwei Wang.

Figure 1
Figure 1. Figure 1: A schematic diagram of a 5 × 5 planar code. Data qubits, X-measurement qubits, and Z-measurement qubits are represented by gray, green, and yellow circles, respectively. The green and yellow lines indicate that the corresponding X or Z stabilizer measurements involve applying X or Z operators to their adjacent data qubits, respectively. In (a), an X-measurement qubit couples to its four adjacent qubits. In… view at source ↗
Figure 2
Figure 2. Figure 2: A schematic diagram of a 5 × 5 XYZ planar code. Data qubits, X-measurement qubits, and ZY - measurement qubits are represented by gray, green, and yellow circles, respectively. The green and yellow lines indicate that the corresponding X or ZY stabilizer measurements involve applying X or Z operators to their adjacent data qubits, respectively, while the purple lines indicate that the ZY stabilizer measure… view at source ↗
Figure 3
Figure 3. Figure 3: The structure of all ZY stabilizers in the XYZ planar code can be obtained by concatenating subfigures (a) and (b). In both subfigures (a) and (b), qubit 4 is simultaneously involved in the Y operators of two ZY stabilizers, and is therefore referred to as the central qubit, while the remaining qubits are called peripheral qubits. non-trivially to the measurement outcome, and “no flip” otherwise. For examp… view at source ↗
Figure 4
Figure 4. Figure 4: Logical failure rate f as a function of the rescaled error rate x = (p − pc)d 1/ν for biases η ∈ {1, 10, 100, 1000}. The solid line is the best fit to the model f = A + Bx + Cx2 . The insets show the raw sample means over 100,000 runs for various values of p. Good agreement between the fitted model and the data is observed across all bias cases. verify the effectiveness of the decoding strategy, Monte Carl… view at source ↗
Figure 5
Figure 5. Figure 5: Under three different scenarios with the [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Logical failure rate as a function of Z-error bias for a physical error probability of 10%, for the XYZ planar code and the standard planar code with distances 11, 15, 19, and 23. The red curve represents the error probability of the standard planar code decoded using the MWPM decoder, while the blue curve shows the performance of the XYZ planar code decoded using the pMWPM method. 7 SIMULATIONS All simula… view at source ↗
read the original abstract

The minimum-weight perfect matching (MWPM) decoder is a standard decoding strategy for surface codes, but its performance degrades considerably under biased noise. In this paper, a modified surface code, termed the XYZ planar code, is introduced, and the MWPM decoder is extended to posterior MWPM (pMWPM) with almost no increase in decoding complexity. The XYZ planar code exhibits higher and more stable thresholds than the planar code under almost all bias conditions, while also achieving significantly lower logical error rates. Specifically, in the infinite-bias case, the threshold of the XYZ planar code is improved by about \(36\%\) compared to that of the surface code, and it maintains comparable or higher thresholds under other biases -- for example, the threshold reaches approximately \(15.5\%\) at bias \(\eta = 1\) and \(14.2\%\) at \(\eta = 100\). Furthermore, pMWPM can be adapted to a wide range of modified surface codes, and the results presented in this work also indicate its excellent potential in other scenarios, such as configurations in which \(Y\) operators involve a larger number of data qubits.

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

0 major / 2 minor

Summary. The paper introduces the XYZ planar code, a modified surface code, and extends the standard MWPM decoder to posterior MWPM (pMWPM) that incorporates posterior probabilities under the independent biased Pauli noise model (parameterized by bias η) with only local weight updates on the same matching graph. It claims that this combination produces higher and more stable thresholds than the planar/surface code under almost all biases, with a ~36% threshold improvement in the infinite-bias limit and concrete values of ~15.5% at η=1 and ~14.2% at η=100, plus lower logical error rates; the approach is also suggested to generalize to other modified surface codes.

Significance. If the Monte Carlo threshold estimates hold, the work supplies a practical, low-complexity decoder enhancement for biased noise, which is relevant to fault-tolerant quantum computation. Strengths include the clearly defined code construction (Section 3), the explicit posterior-weighted matching procedure (Section 4) that reuses the existing graph structure, and internal consistency of the reported thresholds with the assumed noise model; these elements support the “almost no increase in complexity” claim and the potential for broader applicability.

minor comments (2)
  1. [Abstract] Abstract: the specific numerical thresholds (36% improvement, 15.5% at η=1, 14.2% at η=100) are stated without any reference to lattice sizes, number of samples, or error bars, which would allow readers to assess the Monte Carlo results immediately.
  2. [Section 4] Section 4: while the local weight-update mechanism for pMWPM is described, an explicit small-scale example or pseudocode would clarify how the posterior information is folded into the matching weights without altering the graph topology.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive assessment of our work on the XYZ planar code and pMWPM decoder, including the noted strengths in code construction, the posterior-weighted matching procedure, and internal consistency of the thresholds. The recommendation for minor revision is appreciated. No specific major comments are listed in the report.

Circularity Check

0 steps flagged

No significant circularity; thresholds from explicit Monte Carlo simulation

full rationale

The paper defines the XYZ planar code construction (Section 3) and pMWPM decoder (Section 4) explicitly from the independent biased Pauli noise model parameterized by η. Reported thresholds (e.g., 36% infinite-bias improvement, 15.5% at η=1) are obtained via Monte Carlo sampling of logical error rates; these are not fitted parameters, self-definitions, or renamings of inputs. No load-bearing self-citations or ansatzes are invoked to force the central performance claims. The derivation chain is self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 2 invented entities

The central claim rests on the newly defined XYZ planar code lattice and the pMWPM decoder rule, both introduced in the paper, plus the standard assumption of independent biased Pauli noise.

axioms (1)
  • domain assumption Independent identically distributed Pauli errors with bias parameter η.
    Implicit background assumption of all surface-code threshold studies.
invented entities (2)
  • XYZ planar code no independent evidence
    purpose: Modified surface code lattice intended to improve biased-noise performance
    Newly introduced construction whose definition is internal to the paper.
  • posterior MWPM (pMWPM) no independent evidence
    purpose: Decoder that incorporates posterior probabilities into MWPM matching
    New extension proposed in the paper.

pith-pipeline@v0.9.0 · 5725 in / 1419 out tokens · 40339 ms · 2026-05-25T03:13:26.173776+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

37 extracted references · 37 canonical work pages · 1 internal anchor

  1. [1]

    A. Yu. Kitaev, Fault-tolerant quantum computation by anyons, Ann. Phys.303, 2 (2003)

    work page 2003

  2. [2]

    S. B. Bravyi and A. Yu. Kitaev, Quantum codes on a lattice with boundary, arXiv:quant-ph/9811052 (1998)

    work page internal anchor Pith review Pith/arXiv arXiv 1998

  3. [3]

    Aliferis, F

    P. Aliferis, F. Brito, D. P. DiVincenzo, J. Preskill, M. Steffen, and B. M. Terhal, Fault-tolerant computing with biased-noise superconducting qubits: A case study, New J. Phys.11, 013061 (2009). 13 Table 2: Conditional probabilities and weights forη= 10. n= 2 n= 3 p s1 ̸=s 2 s1 =s 2 = 0s 1 =s 2 = 1 s1 ̸=s 2 s1 =s 2 = 0s 1 =s 2 = 1 0.02 0.019 091 0.000 92...

    work page 2009

  4. [4]

    M. D. Shulman, O. E. Dial, S. P. Harvey, H. Bluhm, V. Umansky, and A. Yacoby, Demonstration of entanglement of electrostatically coupled singlet-triplet qubits, Science336, 202 (2012)

    work page 2012

  5. [5]

    D. Nigg, M. M¨ uller, E. A. Martinez, P. Schindler, M. Hennrich, T. Monz, M. A. Martin-Delgado, and R. Blatt, Quantum computations on a topologically encoded qubit, Science345, 302 (2014)

    work page 2014

  6. [6]

    D. K. Tuckett, S. D. Bartlett, and S. T. Flammia, Ultrahigh error threshold for surface codes with biased noise, Phys. Rev. Lett.120, 050505 (2018)

    work page 2018

  7. [7]

    D. K. Tuckett, A. S. Darmawan, C. T. Chubb, S. Bravyi, S. D. Bartlett, and S. T. Flammia, Tailoring surface codes for highly biased noise, Phys. Rev. X9, 041031 (2019)

    work page 2019

  8. [8]

    J. P. Bonilla Ataides, D. K. Tuckett, S. D. Bartlett, S. T. Flammia, and J. B. Brown, The XZZX surface code, Nat. Commun.12, 2172 (2021)

    work page 2021

  9. [9]

    A. Dua, A. Kubica, L. Jiang, S. T. Flammia, and M. J. Gullans, Clifford-deformed surface codes, PRX Quantum5, 010347 (2024)

    work page 2024

  10. [10]

    Edmonds, Paths, trees, and flowers, Can

    J. Edmonds, Paths, trees, and flowers, Can. J. Math.17, 449 (1965)

    work page 1965

  11. [11]

    A. G. Fowler, Minimum weight perfect matching of fault-tolerant topological quantum error correc- tion in average O(1) parallel time, Quantum Inf. Comput.15, 145 (2015)

    work page 2015

  12. [12]

    Higgott, PyMatching: A Python package for decoding quantum codes with minimum-weight perfect matching, ACM Trans

    O. Higgott, PyMatching: A Python package for decoding quantum codes with minimum-weight perfect matching, ACM Trans. Quantum Comput.3, 1 (2022)

    work page 2022

  13. [13]

    Higgott and C

    O. Higgott and C. Gidney, Sparse blossom: Correcting a million errors per core second with minimum-weight matching, Quantum9, 1600 (2025)

    work page 2025

  14. [14]

    Wu and L

    Y. Wu and L. Zhong, Fusion blossom: Fast MWPM decoders for QEC, in2023 IEEE Interna- tional Conference on Quantum Computing and Engineering (QCE)(IEEE Computer Society, Los Alamitos, CA, 2023), pp. 928–938

    work page 2023

  15. [15]

    Delfosse and N

    N. Delfosse and N. H. Nickerson, Almost-linear time decoding algorithm for topological codes, Quantum5, 595 (2021)

    work page 2021

  16. [16]

    Higgott, T

    O. Higgott, T. C. Bohdanowicz, A. Kubica, S. T. Flammia, and E. T. Campbell, Improved decoding of circuit noise and fragile boundaries of tailored surface codes, Phys. Rev. X13, 031007 (2023). 15

    work page 2023

  17. [17]

    Panteleev and G

    P. Panteleev and G. Kalachev, Degenerate quantum LDPC codes with good finite length perfor- mance, Quantum5, 585 (2021)

    work page 2021

  18. [18]

    Roffe, D

    J. Roffe, D. R. White, S. Burton, and E. T. Campbell, Decoding across the quantum low-density parity-check code landscape, Phys. Rev. Res.2, 043423 (2020)

    work page 2020

  19. [19]

    Bravyi, M

    S. Bravyi, M. Suchara, and A. Vargo, Efficient algorithms for maximum likelihood decoding in the surface code, Phys. Rev. A90, 032326 (2014)

    work page 2014

  20. [20]

    Herold, E

    M. Herold, E. T. Campbell, J. Eisert, and M. J. Kastoryano, Cellular-automaton decoders for topological quantum memories, npj Quantum Inf.1, 15010 (2015)

    work page 2015

  21. [21]

    Kubica and J

    A. Kubica and J. Preskill, Cellular-automaton decoders with provable thresholds for topological codes, Phys. Rev. Lett.123, 020501 (2019)

    work page 2019

  22. [22]

    Duclos-Cianci and D

    G. Duclos-Cianci and D. Poulin, Fast decoders for topological quantum codes, Phys. Rev. Lett.104, 050504 (2010)

    work page 2010

  23. [23]

    Chamberland and P

    C. Chamberland and P. Ronagh, Deep neural decoders for near term fault-tolerant experiments, Quantum Sci. Technol.3, 044002 (2018)

    work page 2018

  24. [24]

    Varsamopoulos, B

    S. Varsamopoulos, B. Criger, and K. Bertels, Decoding small surface codes with feedforward neural networks, Quantum Sci. Technol.3, 015004 (2018)

    work page 2018

  25. [25]

    Berent, L

    L. Berent, L. Burgholzer, P. J. H. S. Derks, J. Eisert, and R. Wille, Decoding quantum color codes with MaxSAT, Quantum8, 1506 (2024)

    work page 2024

  26. [26]

    deMarti iOlius, J

    A. deMarti iOlius, J. Etxezarreta Martinez, P. Fuentes, P. M. Crespo, and J. Garcia-Fr´ ıas, Perfor- mance of surface codes in realistic quantum hardware, Phys. Rev. A106, 062428 (2022)

    work page 2022

  27. [27]

    deMarti iOlius, J

    A. deMarti iOlius, J. Etxezarreta Martinez, P. Fuentes, and P. M. Crespo, Performance enhancement of surface codes via recursive minimum-weight perfect-match decoding, Phys. Rev. A108, 022401 (2023)

    work page 2023

  28. [28]

    Dennis, A

    E. Dennis, A. Kitaev, A. Landahl, and J. Preskill, Topological quantum memory, J. Math. Phys. 43, 4452 (2002)

    work page 2002

  29. [29]

    B. M. Terhal, Quantum error correction for quantum memories, Rev. Mod. Phys.87, 307 (2015)

    work page 2015

  30. [30]

    A. G. Fowler, M. Mariantoni, J. M. Martinis, and A. N. Cleland, Surface codes: Towards practical large-scale quantum computation, Phys. Rev. A86, 032324 (2012)

    work page 2012

  31. [31]

    Padmanabhan, A

    P. Padmanabhan, A. Chowdhury, F. Sugino, M. Guha Majumdar, and K. K. Sabapathy, Non-CSS color codes on 2D lattices: Models and topological properties, arXiv:2112.13617 (2021)

    work page arXiv 2021

  32. [32]

    D. S. Wang, A. G. Fowler, A. M. Stephens, and L. C. L. Hollenberg, Threshold error rates for the toric and planar codes, Quantum Inf. Comput.10, 456 (2010)

    work page 2010

  33. [33]

    deMarti iOlius, P

    A. deMarti iOlius, P. Fuentes, R. Or´ us, and P. M. Crespo, Review on the decoding algorithms for surface codes, Quantum8, 1180 (2024)

    work page 2024

  34. [34]

    C. Wang, J. Harrington, and J. Preskill, Confinement-Higgs transition in a disordered gauge theory and the accuracy threshold for quantum memory, Ann. Phys.303, 31 (2003)

    work page 2003

  35. [35]

    F. H. E. Watson and S. D. Barrett, Logical failure rate scaling of the toric code, New J. Phys.16, 093045 (2014)

    work page 2014

  36. [36]

    Y. Xiao, B. Srivastava, and M. Granath, Exact results on finite size corrections for surface codes tailored to biased noise, Quantum8, 1468 (2024)

    work page 2024

  37. [37]

    Delfosse and J

    N. Delfosse and J. P. Tillich, A decoding algorithm for CSS codes using the X/Z correlations, in2014 IEEE International Symposium on Information Theory(IEEE, New York, 2014), pp. 1071–1075. 16

    work page 2014