REVIEW 3 major objections 7 minor 31 references
Elucidating the Physical and Mathematical Properties of the Prouhet-Thue-Morse Sequence in Quantum Computing
T0 review · 3 major / 7 minor · reviewed 2026-08-10 · deepseek-v4-flash
Pith's one-line read Prouhet–Thue–Morse states form a phase-flip error-correcting code and a noise-resistant memory in X-X Ising chains.
desk verdict Correct but mostly a rename: the PTM states are the standard parity GHZ states, and the paper's own dephasing model undercuts its central robust-memory claim. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing object is the PTM logical-state pair, $|0_{\mathrm{TM}}^{(N)}\rangle \propto \sum_{e\in E(N)} |e\rangle$ and $|1_{\mathrm{TM}}^{(N)}\rangle \propto \sum_{o\in O(N)} |o\rangle$, where $E(N)$ and $O(N)$ split the first $2^N$ indices by the parity of their binary digit sum. The equal-power-sum identity between these two index sets, the Prouhet–Tarry–Escott property, is what makes the diagonal error expectations equal, and the fact that flipping two qubits preserves the parity class is what makes the states eigenstates of X-X rotations. The Hadamard transform converts $|0\ldots0\rangle$ and $|1\ldots1\rangle$ into symmetric and antisymmetric combinations of the two PTM states, giving a preparation circuit, while the Lindblad evolution under $S_z$ dephasing preserves each parity class, so the PTM bit functions as the indicator of which mixture survives.
What would settle it
Prepare $|0_{\mathrm{TM}}^{(N)}\rangle$ in an X-X Ising chain, inject a single bit-flip or amplitude-damping error, and compare the decoded logical fidelity with the no-error case; if it decays like an unencoded qubit, the claimed protection is absent. Alternatively, compute the Knill–Laflamme matrix element for a product of all $N$ Pauli $\sigma_z$ operators: if the diagonal entries for $|0_{\mathrm{TM}}^{(N)}\rangle$ and $|1_{\mathrm{TM}}^{(N)}\rangle$ differ, the protection stops below $N$ phase flips exactly as the paper's $M<N$ condition states.
Extended reading notes
Core claim
The central claim is that the two PTM logical states $|0_{\mathrm{TM}}^{(N)}\rangle$ and $|1_{\mathrm{TM}}^{(N)}\rangle$, built from the equal-power-sum split of the first $2^N$ Thue–Morse indices, form a protected code subspace in an X-X Ising spin chain. For any product of fewer than $N$ Pauli $\sigma_z$ operators on distinct qubits, the diagonal matrix elements for the two states agree and the off-diagonal elements vanish (Property 1.2.2); the paper identifies this as the Knill–Laflamme condition, so up to $(N-1)/2$ single-qubit phase-flip errors are detectable and, given ancillas, correctable. Every X-X rotation $e^{i\theta \sigma_x^{(k)}\sigma_x^{(j)}}$ leaves each PTM state unchanged up to the same global phase, making the encoded subspace transparent to the X-X Ising Hamiltonian. The two states are also exchanged by $S_x$, the Hadamard transform maps them from simple computational-basis superpositions, and the same Prouhet–Tarry–Escott identity extends the construction to qudits.
Load-bearing premise
The entire memory-protection and error-correction case assumes the only noise acting on each qubit is pure phase-flip dephasing along $z$; if bit-flip or amplitude-damping errors are present the PTM states are not shown to be protected, and the zeta measurement additionally assumes that power-law superposition states can be prepared in a logarithmic-spectrum oscillator.
Editorial extensions
If this is right
- Encoding one logical qubit as $\alpha|0_{\mathrm{TM}}^{(N)}\rangle + \beta|1_{\mathrm{TM}}^{(N)}\rangle$ in an X-X Ising chain with pure dephasing preserves the logical information, because all X-X couplings contribute only a global phase.
- Phase-flip errors on up to $(N-1)/2$ qubits are detectable and, with ancillas, correctable, because the PTM states meet the Knill–Laflamme conditions.
- The PTM encoding removes first-order sensitivity to uniform external fields along any axis, and superposition states are additionally insensitive to $y$ and $z$ magnetic-field noise.
- Applying the quantum Fourier transform to a PTM state yields self-similar, multifractal amplitude profiles, linking the sequence to approximate eigenstates of the quantum baker's map.
- The PTM-weighted Dirichlet-series identity for the Riemann zeta function suggests a two-state interference measurement in a logarithmic-spectrum oscillator whose autocorrelation returns $\zeta(s)$.
Reading between the lines
- The protection is tied to the dephasing channel $L_k \propto S_z^{(k)}$; simulating bit-flip or amplitude-damping noise on the same encoding would likely show rapid loss of logical fidelity, since the proof uses only diagonality of $\sigma_z$.
- Because every pair of $\sigma_x$ flips stabilizes the code space, concatenating the PTM encoding with a stabilizer code that handles $X$ errors might cover both error types, but the paper does not analyze such a concatenation.
- The qudit version of the construction suggests a family of high-dimensional encodings whose noise-detection order grows with $\log_2 d$, but no decoding circuit or resource count is given.
- The proposed $\zeta(s)$ measurement could be benchmarked classically for small $N$ by simulating the stated superposition states and checking that the computed autocorrelation matches $\zeta(s)$ to the expected precision.
Formalized claims in Lean
-
Claim #1: The central claim is that the two PTM logical states $|0_{\mathrm{TM}}^{(N)}\rangle$ and $|1_{\mathrm{TM}}^{(N)}\rangle$, built from the equal-power-sum split of the first $2^N$ Thue–Morse indices, form a protected code subspace in an X-X Ising spin chain. For any product of fewer than $N$ Pauli $\sigma_z$ operators on distinct qubits, the diagonal matrix elements for the two states agree and the
/-- @claim 1 The central claim is that the two PTM logical states $|0_{\mathrm{TM}}^{(N)}\rangle$ and $|1_{\mathrm{TM}}^{(N)}\rangle$, built from the equal-power-sum split of the first $2^N$ Thue–Morse indices, form a protected code subspace in an X-X Ising spin chain. For any product of fewer than $N$ Pauli $\sigma_z$ operators on distinct qubits, the diagonal matrix elements for the two states agree and the -/ def central_claim : Prop :=
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper introduces two logical states, |0_TM^(N)> and |1_TM^(N)>, formed by equal-amplitude superpositions over the even-parity (E(N)) and odd-parity (O(N)) computational basis states determined by the Prouhet-Thue-Morse sequence. It proves several algebraic properties of these states: vanishing expectation values of total spin operators, Knill-Laflamme-type conditions for products of σ_z operators, and invariance under pair flips σ_x^(k)σ_x^(j). These properties are then used to discuss quantum error correction, a proposed noise-resistant quantum memory in X-X Ising chains, connections to the Walsh-Hadamard transform and the quantum baker's map, and a scheme to measure the Riemann zeta function via states with power-law coefficients in a logarithmic-spectrum oscillator. The derivations are elementary and mostly correct; the central concern is that the advertised applications, especially the robust-memory claim, are not supported by the paper's own open-system analysis.
Significance. If the results were fully established, the paper would offer a compact encoding perspective on phase-flip error correction and on the symmetries of X-X Ising chains. The algebraic core is correct: Eq. (33) correctly describes the QFT amplitudes of the difference state, and Property 1.2.2 is a valid verification of the Knill-Laflamme conditions for products of fewer than N σ_z operators. The paper is self-contained and the proofs in Appendix B are checkable. However, the significance is limited by three issues: the QEC code is the standard phase-flip code, the robust-memory claim is contradicted by the paper's own dephasing model, and the zeta-function proposal is a restatement of a Dirichlet-series identity without a feasibility analysis. These overstatements affect the abstract and the conclusion, not just the presentation.
major comments (3)
- [Sec. 2.1, Eq. (23); Sec. 3.2, Eq. (26)] The 'robust encoding of quantum memories' claim is not supported by the paper's own noise model. The master equation (23) contains local dephasing with L_k = S_z^(k); for a logical superposition α|0_TM> + β|1_TM>, the off-diagonal coherences between E(N) and O(N) basis states decay on a timescale ~1/γ, so the logical coherence <0_TM|ρ|1_TM> is lost. The paper itself states in Section 2.1 that 'if the initial state includes eigenstates from both sets, the final state will be a mixture of all eigenstates.' Property 1.2.1 and the matrix representations in Eq. (26) only establish first-order insensitivity to global magnetic fields, not robustness to the Lindblad dephasing introduced in Eq. (23). The invariance under X-X rotations (Property 1.2.4, Section 1.4) does not protect the memory because H_X does not commute with L_k. The abstract's 'robust encoding' should be withdrawn or explicitly restricted to closed-system Hamiltonian perturbations, or an active error-correction analysis for the memory setting must be supplied.
- [Sec. 3.1, Property 1.2.2] The error-correction claim is misstated. Property 1.2.2 verifies the Knill-Laflamme conditions for every product of fewer than N σ_z operators. This means that up to N-1 single-qubit phase-flip errors are detectable, and, because E_a†E_b for two errors of weight ≤ t is a product of at most 2t σ_z operators, up to floor((N-1)/2) errors are correctable. The sentence 'up to (N-1)/2 single-qubit phase flip errors are detectable' should read 'up to floor((N-1)/2) errors are correctable, and up to N-1 errors are detectable.' The distinction matters because the N=3 circuit in Fig. 5 corrects one error, which matches the correctability bound, but the stated detectability bound of (N-1)/2 would incorrectly suggest that only one error can be detected for N=3.
- [Sec. 3.4, Eqs. (37)-(38)] The zeta-function measurement is not demonstrated as a feasible quantum protocol. The proposal relies on preparing the infinite superpositions |ψ1> and |ψ2> in Eq. (37), with amplitudes t_n (n+1)^{-σ/2}, in a logarithmic-spectrum oscillator, and no truncation, state-preparation, or measurement-error analysis is given. The 'exact value of ζ(s)' obtained from the conjugate autocorrelation is a restatement of the Dirichlet-series identity ζ(s)=(1+1/2^s)Σ_{n≥1} t_{n-1}/n^s + (1-1/2^s)Σ_{n≥1} t_n/n^s, not a new algorithmic result. If the claim is only an in-principle correspondence, the text should say so explicitly; otherwise the resource requirements and finite-dimensional truncation errors should be analyzed.
minor comments (7)
- [Sec. 1.2, Property 1.2.4] The index range '∀k, j < N' should be '1 ≤ k,j ≤ N' for consistency with the notation in Eq. (17).
- [Sec. 1.5] The title 'PTM states as eigenvalues of Sx' should read '... as eigenstates of Sx'; the states are eigenstates, not eigenvalues.
- [Sec. 3.3, Eq. (27)] The index range in the definition of the QFT gate is stated as '0 ≤ j,k < N'; since the gate acts on N qubits, the correct range is 0 ≤ j,k < 2^N.
- [Fig. 3 caption] The initial states are written as |ψ±(0)> = 1/√2 (|(0)_2> ± |(N)_2>), but the context of Eq. (24) indicates the second state should be |(2^N - 1)_2>, the all-ones state.
- [Sec. 3.1, Fig. 4] The caption calls the circuit 'Shor's 3-qubit phase-flip error correction code'; the 3-qubit phase-flip code is distinct from Shor's 9-qubit code, so the attribution should be corrected to avoid confusion.
- [Sec. 3.1, Eq. (25)] The qudit generalization should explicitly restrict d to a power of two (d = 2^N) before defining the PTM states with the prefactor sqrt(2/d), since the equal-size property of E(N) and O(N) used for normalization holds only in that case.
- [Conclusion] The conclusion states that the logical states exhibit 'resilience to spin flip errors'; the supported property is resilience to phase-flip (σ_z) errors, not spin-flip (σ_x) errors, and the wording should be amended.
Circularity Check
No significant circularity: PTM properties are derived from definitions; self-citations are peripheral.
full rationale
The paper's derivation chain is self-contained. The PTM states are defined directly from the PTM sequence in Eqs. (15)-(16), and Properties 1.2.1-1.2.4, including the Knill-Laflamme conditions, are proven by induction in Appendix B from those definitions and the Prouhet-Tarry-Escott identity of Eq. (7), not imported from fitted parameters or from the cited literature. The apparent 'appearance' of the PTM sequence in the dephased X-X Ising chain (Section 2.1) is a direct mathematical consequence of the dephasing master equation preserving the parity subspaces E(N) and O(N) that define the PTM states; no parameter is fitted and no prediction is forced by construction. The zeta-function protocol (Section 3.4) restates the quoted Dirichlet-series identity for the Riemann zeta function and combines it with the Feiler-Schleich overlap formula: the measured autocorrelation amplitude is proportional to the Dirichlet series by construction, and the series equals zeta(s) by the quoted identity. This is a derived equivalence rather than a circular input. The only self-citations, [13,14], support a peripheral remark about qudit error rates and decoherence times and play no role in deriving the PTM properties; no load-bearing argument reduces to a self-citation. The memory-robustness concern raised by the skeptic is a physical-correctness issue about dephasing of superpositions, not a circularity in the derivation chain.
Assumptions & free parameters
assumptions (4)
- standard math The PTM sequence properties used in the paper, including the Prouhet-Tarry-Escott identity (Eq. 7) and the product formula (Eq. 30), are assumed as standard results.
- domain assumption The Lindblad master equation (Eq. 23) with dephasing operators L_k = S_z^(k) is taken as the noise model for the X-X Ising chain.
- standard math The Knill-Laflamme conditions are used as the criterion for quantum error correction.
- domain assumption The Feiler-Schleich oscillator with logarithmic energy spectrum (H = omega ln(n+1)|n><n|) is assumed to be realizable and the states of Eq. (37) preparable.
Cite this review
Pith. "Pith review of Elucidating the Physical and Mathematical Properties of the Prouhet-Thue-Morse Sequence in Quantum Computing." pith.science (2026). https://pith.science/paper/MFFR4EXA
@misc{pith2026250109610,
author = {Pith},
title = {Pith review of: Elucidating the Physical and Mathematical Properties of the Prouhet-Thue-Morse Sequence in Quantum Computing},
year = {2026},
howpublished = {\url{https://pith.science/paper/MFFR4EXA}},
note = {Machine review of arXiv:2501.09610}
}
read the original abstract
This study explores the applications of the Prouhet-Thue-Morse (PTM) sequence in quantum computing, highlighting its mathematical elegance and practical relevance. We demonstrate the critical role of the PTM sequence in quantum error correction, in noise-resistant quantum memories, and in providing insights into quantum chaos. Notably, we demonstrate how the PTM sequence naturally appears in Ising X-X interacting systems, leading to a proposed robust encoding of quantum memories in such systems. Furthermore, connections to number theory, including the Riemann zeta function, bridge quantum computing with pure mathematics. Our findings emphasize the PTM sequence's importance in understanding the mathematical structure of quantum computing systems and the development of the full potential of quantum technologies and invite further interdisciplinary research.
Figures
Figures from the paper (4 more)
Reference graph
Works this paper leans on
-
[1]
Introduction The intersection of quantum computing and mathematics opens a fascinating frontier for research, with potential implications for both theoretical advancements and practical applications. Among mathematical constructs, the Prouhet-Thue-Morse (PTM) sequence stands out because of its surprisingly rich structure and multifaceted utility. Original...
work page 1906
-
[2]
Appearance of the PTM sequence in the Hilbert spaces of given quantum computing platforms 2.1. The PTM sequence as the indicator function of the purely dephased X-X Ising chain Definition: We define an X-X Ising chain of decohering qubits as a system of N qubits evolving according to the equation: Elucidating the Physical and Mathematical Properties of th...
-
[3]
Uses of the PTM sequence in quantum computing 3.1. Quantum Error Correction (QEC) Property 1.2.2 shows that the PTM states satisfy the Knill-Laflamme conditions [12], meaning up to N −1 2 single-qubit phase flip errors are detectable, and provided additional ancilla qubits, correctable. Let us consider the traditional 3 qubit phase-flip error detection ci...
work page 2020
-
[4]
Conclusion In this study, we have explored the many manifestations and uses of the Prouhet-Thue- Morse (PTM) sequence in quantum computing, highlighting its intrinsic mathematical interest and practical significance. Through rigorous analysis, we have shown that the PTM sequence, beyond its mathematical appeal, plays a central role in quantum error correc...
-
[5]
Acknoledgments This work was funded by the French National Research Agency (ANR) through the Programme d’Investissement d’Avenir under contract ANR-11-LABX-0058 NIE and ANR-17-EURE-0024 within the Investissement d’Avenir program ANR-10-IDEX-0002-
-
[6]
Specifically, the (PTM) sequence A010060
The authors would like to thank the Online Encyclopedia of Integer Sequences (OEIS) for providing valuable data and resources that contributed to this research. Specifically, the (PTM) sequence A010060
-
[7]
Data availability The codes and calculations done in this study are available from the authors upon reasonable request
-
[8]
Conflicts of interest The authors declare no conflicts of interests
Show all 31 references
-
[9]
Prouhet E 1851 C. R. Acad. Sci. Paris Ser. I33 225
-
[10]
Allouche J P and Shallit J 1999 The ubiquitous prouhet-thue-morse sequence Sequences and their Applications ed Ding C, Helleseth T and Niederreiter H (London: Springer London) pp 1–16 ISBN 978-1-4471-0551-0
1999
-
[11]
Kol´ aˇ r M, Ali M K and Nori F 1991Phys. Rev. B43(1) 1034–1047 URL https://link.aps.org/ doi/10.1103/PhysRevB.43.1034
-
[12]
Thue A 1906 Kra. Vidensk. Selsk. Skrifter. I. Mat.-Nat. Kl
1906
-
[13]
semanticscholar.org/CorpusID:54043171
Morse H M Transactions of the American Mathematical Society22 84–100 URL https://api. semanticscholar.org/CorpusID:54043171
-
[14]
Xiong L, Zhang Y, Liu Y, Zheng Y and Jiang X 2022 Physical Review Applied18 064089 URL https://link.aps.org/doi/10.1103/PhysRevApplied.18.064089 Elucidating the Physical and Mathematical Properties of the PTM Sequence in QC 18
2022 doi
-
[15]
Deng X H, ji Ren Yuan, Hong W Q and Ouyang H 2011 Physics Procedia22 360–365 ISSN 1875- 3892 2011 International Conference on Physics Science and Technology (ICPST 2011) URL https://www.sciencedirect.com/science/article/pii/S1875389211007103
2011
-
[16]
Matarazzo V, De Nicola S, Zito G, Mormile P, Rippa M, Abbate G, Zhou J and Petti L 2010 Journal of Optics13 015602 ISSN 2040-8986 URL http://dx.doi.org/10.1088/2040-8978/ 13/1/015602
2010 doi
-
[17]
Yang J K, Noh H, Boriskina S V, Rooks M J, Solomon G S, Dal Negro L and Cao H 2012 Lasing in thue-morse structure with optimal aperiodicity 2012 Conference on Lasers and Electro-Optics (CLEO) pp 1–2
2012
-
[18]
Nguyen H D 2014 A new proof of the prouhet-tarry-escott problem URL https://arxiv.org/ abs/1411.6168
2014 arXiv
-
[19]
Aifer M and Deffner S 2022 New Journal of Physics 24 055002 ISSN 1367-2630 URL http: //dx.doi.org/10.1088/1367-2630/ac6821
2022 doi
-
[20]
Knill E and Laflamme R 1997 Phys. Rev. A 55(2) 900–911 URL https://link.aps.org/doi/ 10.1103/PhysRevA.55.900
1997 doi
-
[21]
Jankovi´ c D, Hartmann J G, Ruben M and Hervieux P A 2024npj Quantum Information10 ISSN 2056-6387 URL http://dx.doi.org/10.1038/s41534-024-00829-6
-
[22]
Hartmann J G, Jankovi´ c D, Pasquier R, Ruben M and Hervieux P A 2024 Nonlinearity of the fidelity in open qudit systems: Gate and noise dependence in high-dimensional quantum computing URL https://arxiv.org/abs/2406.15141
2024 arXiv
-
[23]
Chiesa A, Macaluso E, Petiziol F, Wimberger S, Santini P and Carretta S 2020 The Journal of Physical Chemistry Letters11 8610–8615 ISSN 1948-7185 URL http://dx.doi.org/10.1021/ acs.jpclett.0c02213
2020
-
[24]
Meenakshisundaram N and Lakshminarayan A 2005 Phys. Rev. E 71(6) 065303 URL https: //link.aps.org/doi/10.1103/PhysRevE.71.065303
2005 doi
-
[25]
Maity K and Lakshminarayan A 2006 Phys. Rev. E74(3) 035203 URL https://link.aps.org/ doi/10.1103/PhysRevE.74.035203
2006 doi
-
[26]
doi.org/10.1137/S0097539795293172
Shor P W 1997 SIAM Journal on Computing 26 1484–1509 ISSN 1095-7111 URL http://dx. doi.org/10.1137/S0097539795293172
1997 doi
-
[27]
Fan A, Schmeling J and Shen W 2022 Multifractal analysis of generalized thue-morse trigonometric polynomials URL https://arxiv.org/abs/2212.13234
2022 arXiv
-
[28]
T´ oth L 2022 URLhttps://arxiv.org/abs/2211.13570
2022 arXiv
-
[29]
Feiler C and Schleich W P 2015 New Journal of Physics 17 063040 ISSN 1367-2630 URL http://dx.doi.org/10.1088/1367-2630/17/6/063040
2015 doi
-
[30]
Gleisberg F, Mack R, Vogel K and Schleich W P 2013 New Journal of Physics15 023037 ISSN 1367-2630 URL http://dx.doi.org/10.1088/1367-2630/15/2/023037
2013 doi
-
[31]
Notations Hilbert Space: A Hilbert Space is a finite-dimensional ( d) vector space over C, equipped with a sesquilinear inner product ⟨·|·⟩
Feiler C and Schleich W P 2013 New Journal of Physics 15 063009 ISSN 1367-2630 URL http://dx.doi.org/10.1088/1367-2630/15/6/063009 Appendix A. Notations Hilbert Space: A Hilbert Space is a finite-dimensional ( d) vector space over C, equipped with a sesquilinear inner product ...
2013 doi
Reviewed August 10, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.