Pith. sign in

REVIEW 3 major objections 3 minor 21 references

Modular Debiasing: A Robust Method for Quantum Randomness Extraction

T0 review · 3 major / 3 minor · reviewed 2026-08-16 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read For any m-sided source in which every outcome can occur, summing N draws modulo m drives the output distribution to uniform with error bounded by ((m-1)/m) rho^N, and the guarantee persists when probabilities drift across trials.

desk verdict The IID modular-sum debiasing result is correct but standard, while the advertised non-stationary robustness theorem is false as stated and the novelty claim is unsupported. read the letter →

arxiv 2504.18585 v2 pith:FMLF6YB3 submitted 2025-04-23 physics.data-an cs.ITmath.ITmath.PRquant-ph

Eduardo Gueron This is my paper · ORCID
classification physics.data-ancs.ITmath.ITmath.PRquant-ph MSC 60E10
keywords quantumrandomnumbergeneratorrandomnessextractiondebiasingmodulararithmeticprobabilitygeneratingfunctionsrootsofunitynon-stationarysourcesspatialphotondetection
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper aims to prove that one arithmetic operation—adding the outcomes of N independent draws from a biased m-sided source and reducing the sum modulo m—turns the output into an almost perfectly uniform distribution over {0, 1, ..., m-1}. The argument uses probability generating functions and roots of unity: the exact probability of each residue is 1/m plus a finite sum of contractive factors, each raised to the N-th power. Full support (every outcome has positive probability) forces every nonconstant Fourier coefficient to have magnitude strictly below 1, so the error decays exponentially with N. The paper further claims that convergence survives non-stationary sources whose probabilities change from trial to trial, provided each trial keeps every outcome possible. If correct, this gives spatial-photon quantum random number generators a trivial post-processing step that removes detector and illumination bias without discarding any raw events.

What carries the argument

The load-bearing identity is the root-of-unity extraction formula P(r_f ≡ k mod m) = (1/m) Σ_{r=0}^{m-1} $ω^{{-rk}}$ ∏_{j=1}^N G_j(ω^r), where G_j(t) = Σ_k $p^{{(j)}}$_k t^k is the probability generating function of trial j and ω = $e^{{2πi/m}}$. The r = 0 term contributes exactly 1/m; every r ≠ 0 term is a product of Fourier coefficients G_j(ω^r), each of magnitude ≤ 1, and strictly less than 1 under full support. The rate parameter ρ = max_{r≠0} |Σ_l p_l $ω^{{rl}}$| sets the exponential contraction rate. The machinery converts a question about residue statistics of a sum into a question about whether a product of complex numbers on the unit disk contracts to zero.

What would settle it

Run the modular sum for m = 2 with $p^{{(j)}}$(0) = 1 − $2^{{-(j+1)}}$ and $p^{{(j)}}$(1) = $2^{{-(j+1)}}$. The trial product equals ∏_{j=1}^N (1 − $2^{{-j}}$), which converges to a positive constant c ≈ 0.2888, so P(r_f = 0) tends to (1+c)/2 ≈ 0.644 rather than 1/2. This calculation shows that the divergence condition in Section 4 is doing real work; a simulation of this sequence would settle whether the paper's non-stationary theorem is meant to hold under positivity alone.

Watch

Extended reading notes

Core claim

The central discovery is that modular summation is itself a debiasing map. For independent draws with fixed probabilities p_k > 0, the modular sum r_f satisfies P(r_f ≡ k mod m) = (1/m) Σ_{j=0}^{m-1} $ω^{{-jk}}$ (Σ_l p_l $ω^{{jl}}$)^N, and for every j ≠ 0 the base A_j = Σ_l p_l $ω^{{jl}}$ has |A_j| < 1. The deviation from uniform is therefore bounded by ((m-1)/m) ρ^N with ρ = max_{j≠0} |A_j| < 1. The paper extends the statement to time-dependent probabilities, asserting that the product ∏_j G_j(ω^r) vanishes and the output distribution becomes uniform whenever $p^{{(j)}}$_k remains positive for every trial, so arbitrary initial bias and drift both wash out.

Load-bearing premise

The non-stationary robustness claim relies on an extra condition beyond positivity: the per-trial deviations from perfect uniformity must accumulate, in the sense that ∑_j (1 − |G_j(ω^r)|) diverges for each nonzero frequency; without that condition, a trial sequence can keep every outcome positive yet fail to converge to uniform.

Editorial extensions

If this is right

  • A biased quantum source with m spatial detection regions can be debiased by a simple running counter that adds and reduces modulo m, removing the need to calibrate per-detector efficiency or illumination.
  • No raw samples are discarded: every draw contributes to the final block output, so the output rate approaches log2 m bits per block as N grows.
  • The error bound ((m-1)/m) ρ^N gives a direct way to choose N for a target statistical distance to uniform, even for strongly biased sources.
  • Under the stated average-decay condition, slow parameter drift or cyclic bias does not prevent convergence to uniformity; only a persistent failure to make every outcome possible would leave a residual bias.
  • The method is alphabet-agnostic: the same proof works for any discrete source with m outcomes, including non-binary quantum detection maps.

Reading between the lines

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

  • Going beyond the paper, the bound can be inverted into a practical block-size rule: to guarantee |P - 1/m| ≤ ε, take N ≥ ln(ε m/(m-1)) / ln ρ, which an online extractor could compute on the fly.
  • Going beyond the paper, the product structure suggests a natural extension where each trial has its own alphabet size but a common final modulus; convergence would still hold as long as the Fourier factors contract.
  • Going beyond the paper, the method as stated assumes independent trials; whether modular summation remains a debiaser for weakly dependent sources (e.g., fast-mixing Markov chains) is an open testable extension that the current proof does not cover.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 3 minor

Summary. The paper proposes a modular debiasing method for discrete m-ary sources: the outcomes of N independent trials are summed modulo m, and the resulting residue is used as a supposedly uniform output. The authors claim that, provided every outcome has positive probability, the distribution of the modular sum converges exponentially to the uniform distribution, and that the guarantee persists under non-stationary or time-dependent biases. The derivation uses probability generating functions and roots of unity, derives an IID error bound controlled by the spectral radius rho, and validates the claims with numerical simulations for stationary and non-stationary examples. A practical application to spatial photon-detection QRNGs is also discussed.

Significance. The IID part of the paper is both correct and useful: Eq. (9) gives an exact formula for the modular-sum distribution, and the bound in Eq. (17) is a clean exponential convergence statement for any full-support stationary source. The method is conceptually simple and the numerical experiments are extensive, including entropy and TVD convergence tests. However, the paper's headline contribution is the claimed robustness to non-stationary sources, and that claim is false as stated. Since the QRNG motivation in Section 7 and the abstract's central promise rest on this non-stationary robustness, the current manuscript cannot be accepted without a fundamental reformulation and restriction of the theorem.

major comments (3)
  1. [Section 4, Eqs. (11)-(13)] The inference that positivity of p_j(k) forces the product of Fourier factors to vanish is invalid. From |G_j(omega^r)| < 1 for each j it does not follow that the infinite product tends to 0; this requires a summability condition such as sum_j (1 - |G_j(omega^r)|) = infinity. The paper itself states such a sufficient condition in passing, but it is not assumed in the theorem and is not implied by p_j(k) > 0. A concrete counterexample with m=2 is p_j(0) = 1 - 2^{-(j+1)}, p_j(1) = 2^{-(j+1)}. Then G_j(-1) = 1 - 2^{-j}, so P(r_f=0) = 1/2 + (1/2) prod_{j=1}^N (1 - 2^{-j}), whose limit is (1+C)/2 with C = prod_{j=1}^infty (1-2^{-j}) > 0, not 1/2. All hypotheses of Eq. (13) are satisfied, yet the conclusion fails.
  2. [Section 6 (Non-Stationary simulations)] The numerical validation does not test the theorem as stated. The non-stationary scenarios (random fluctuations around a fixed mean and cyclic bias with a dominant outcome) are special processes in which the logarithms of |G_j| have a negative drift, so the missing summability condition happens to hold. They do not probe the failure mode exhibited by the Section 4 counterexample, where the probabilities approach a perfectly biased limit fast enough that the Fourier product saturates at a positive constant. The simulations therefore cannot provide evidence for the claimed general robustness.
  3. [Section 5, last paragraph; Section 7] The claim that the non-stationary convergence rate is 'also exponential, governed by the average behavior of |G_j|' is not proved and, as written, cannot be true without additional assumptions on the rate at which the sequence approaches degeneracy. The QRNG proposal in Section 7 explicitly rests on robustness to detector-efficiency drift and illumination variation; slow drifts can be exactly the kind of process for which the posited theorem fails, so the practical motivation inherits the Section 4 gap.
minor comments (3)
  1. [Section 3] The sentence 'the complex numbers omega^{jl} for l = 0, ..., m-1 are distinct' is inaccurate when gcd(j,m) > 1; the parenthetical that follows correctly notes periodicity, but the wording should be cleaned up. The strict inequality |A_j| < 1 remains valid because not all nonzero terms have a common phase, but this requires a separate argument in the repeated-root case.
  2. [Section 6, TVD bullet list] The last bullet under 'The TVD values at specific N' repeats the entropy text ('Shows an increase from 1.0979 bits at N=1 ... to 2.3218 bits at N=100') instead of reporting the TVD value for the Non-Stationary Cyclic Bias scenario.
  3. [Figure 3] The log-scale y-axis labels appear corrupted (reading like 10^{-203}, 10^{-176}, etc.), and those values are inconsistent with the TVD values quoted for N=100 in the text. Please regenerate the figure or correct the axis formatting.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the modular-debiasing derivation is self-contained, and its conclusion follows from the stated PGF/root-of-unity calculation rather than from fitted inputs or self-citation.

full rationale

The paper's central derivation (Eqs. 3-17) is a direct analytic calculation: the distribution of the modular sum is expressed via roots of unity, and the IID convergence bound is obtained from |A_j|<1. No input parameter is fitted to the target uniformity; the rate parameter rho is computed from the assumed known bias distribution, not from the data being predicted. The only fitted element in the numerical section is the vertical scale of the dashed theoretical lines in Figure 3, which the paper explicitly says is 'scaled to match the empirical TVD at an early N'; this is a harmless visual alignment and does not determine the slope predicted by rho. There are no self-citations, no imported uniqueness theorems, and no ansatz smuggled in via prior work. The non-stationary robustness argument in Section 4 does contain a real mathematical gap (pointwise |G_j(omega^r)|<1 does not imply product decay without a divergence condition on sum(1-|G_j|)), but that is an error of proof validity, not circularity: the conclusion is not assumed in the hypotheses, and the derivation does not reduce to its inputs by construction. Accordingly, the circularity score is 0.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The mathematical derivation itself has no fitted constants; p_k and m are inputs. The only fitted quantity is the vertical scale of the theoretical lines in Figure 3. The assumptions are the standard ones for convolution of independent variables plus the extra, understated summability condition needed for the non-stationary theorem. No invented physical entities are introduced.

free parameters (1)
  • Theory-line scale in Figure 3 = unspecified, matched to empirical TVD at early N
    The dashed theoretical curves use rho^N computed from the known source distribution, but their vertical position is scaled to match the early empirical TVD. This is a fitted plot scale, not part of the derivation.
assumptions (4)
  • domain assumption Trials are independent: X_1,...,X_N are independent random variables.
    Used to factor the PGF in Eq. 4 and Eq. 10; if photon detections have dead-time or afterpulsing correlations, the factorization fails.
  • domain assumption Every outcome has positive probability at every trial, p_j,k>0.
    Needed to prove strict inequality |A_j|<1 and |G_j(omega^r)|<1; the paper acknowledges zero-probability outcomes as future work.
  • ad hoc to paper For non-stationary sources, the tail Fourier products vanish, e.g. sum_j (1-|G_j(omega^r)|)=infinity.
    Section 4 states this as a sufficient condition but Eqs. 12-13 conclude uniform convergence from positivity alone; without it convergence can fail, as shown by an m=2 example with p_j(0)=1-2^{-(j+1)}.
  • standard math Root-of-unity orthogonality and the identity between coefficient sums mod m and evaluations at roots of unity.
    This is the Fourier inversion step behind Eq. 8; it is standard but not proved.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Modular Debiasing: A Robust Method for Quantum Randomness Extraction." pith.science (2026). https://pith.science/paper/FMLF6YB3

@misc{pith2026250418585,
  author       = {Pith},
  title        = {Pith review of: Modular Debiasing: A Robust Method for Quantum Randomness Extraction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FMLF6YB3}},
  note         = {Machine review of arXiv:2504.18585}
}
abstract

We propose a novel modular debiasing technique applicable to any discrete random source, addressing the fundamental challenge of reliably extracting high-quality randomness from inherently imperfect physical processes. The method involves summing the outcomes of multiple independent trials from a biased source and reducing the sum modulo the number of possible outcomes, $m$. We provide a rigorous theoretical framework, utilizing probability generating functions and roots of unity, demonstrating that this simple operation guarantees the exponential convergence of the output distribution to the ideal uniform distribution over $\{0, 1, \dots, m-1\}$. A key theoretical result is the method's remarkable robustness: convergence is proven for any initial bias (provided all outcomes have non-zero probability) and, crucially, is maintained even under non-stationary conditions or time-dependent noise, which are common in physical systems. Analytical bounds quantify this exponential rate of convergence, and are empirically validated by numerical simulations. This technique's simplicity, strong theoretical guarantees, robustness, and data efficiency make it particularly well-suited for practical implementation in quantum settings, such as spatial photon-detection-based Quantum Random Number Generators (QRNGs), offering an efficient method for extracting high-quality randomness resilient to experimental imperfections. This work contributes a valuable tool to the field of Quantum Information Science.

Figures

Figures reproduced from arXiv: 2504.18585 by the authors.

Figure 1
Figure 1. Entropy Convergence vs N • IID Extreme Bias (m=21): Starting from a high TVD of 0.8517 at N = 1, it drops to 1.26 × 10−2 at N = 100. • Non-Stationary Cyclic Bias (m=5): Shows an increase from 1.0979 bits at N = 1 (vs log2 5 ≈ 2.3219 bits) to 2.3218 bits at N = 100. These results confirm the effectiveness of the method in achieving near-uniform distribu￾tions rapidly for both stationary and non-stationary sources. Fi… view at source ↗
Figure 2
Figure 2. TVD Convergence vs N (Linear Scale) 10 1 10 2 10 3 N (Number of rolls per block) 10 203 10 176 10 149 10 122 10 95 10 68 10 41 10 14 Total Variation Distance (TVD) (log scale) TVD Convergence Towards Zero (Log Scale) IID Strong Bias (m=10, p0=0.8) (m=10) Theory ( N, = 0.78) Non-Stationary Random Noise (m=3) (m=3) IID Extreme Bias (m=21, p0=0.9) (m=21) Theory ( N, = 0.90) Non-Stationary Cyclic Bias (m=5) (m=5) [PITH… view at source ↗
Figure 3
Figure 3. TVD Convergence vs N (Log Scale) 11 [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Proportion Evolution for IID Strong Bias (m=10, p0=0.8) [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]
Figure 5
Figure 5. Figure 5: Proportion Evolution for Non-Stationary Random Noise (m=3) [PITH_FULL_IMAGE:figures/full_fig_p013_5.png]
Figure 6
Figure 6. Figure 6: Proportion Evolution for IID Extreme Bias (m=21, p0=0.9) [PITH_FULL_IMAGE:figures/full_fig_p013_6.png]
Figure 7
Figure 7. Figure 7: Proportion Evolution for Non-Stationary Cyclic Bias (m=5) [PITH_FULL_IMAGE:figures/full_fig_p014_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

21 extracted references · 21 canonical work pages

  1. [1]

    Quantum computing in the nisq era and beyond

    John Preskill. “Quantum computing in the nisq era and beyond”. Quantum 2, 79 (2018)

    work page 2018

  2. [2]

    A mathematical theory of communication

    Claude E. Shannon. “A mathematical theory of communication”. Bell System Tech- nical Journal 27, 379–423 (1948)

    work page 1948

  3. [3]

    True random number generation using the spin crossover in lacoo3

    Kyung Seok Woo, Alan Zhang, Allison Arabelo, Timothy D. Brown, Minseong Park, A. Alec Talin, Elliot J. Fuller, Ravindra Singh Bisht, Xiaofeng Qian, Raymundo Ar- royave, Shriram Ramanathan, Luke Thomas, R. Stanley Williams, and Suhas Kumar. “True random number generation using the spin crossover in lacoo3”. Nature Com- munications 15, 4656 (2024)

    work page 2024

  4. [4]

    Recommendation for the entropy sources used for random bit generation

    Elaine Barker and John Kelsey. “Recommendation for the entropy sources used for random bit generation”. NIST Special Publication (2022)

    work page 2022

  5. [5]

    Maximization of extractable randomness in a quantum random- number generator

    Jing Yan Haw, Syed Assad, Andrew Lance, Nelly Ng, Vikram Sharma, Ping Koy Lam, and T. Symul. “Maximization of extractable randomness in a quantum random- number generator”. Physical Review Applied3 (2014)

    work page 2014

  6. [6]

    Certified randomness using a trapped-ion quantum processor

    Minzhao Liu, Ruslan Shaydulin, Pradeep Niroula, Matthew DeCross, Shih-Han Hung, Wen Yu Kon, Enrique Cervero-Martín, Kaushik Chakraborty, Omar Amer, Scott Aaronson, Atithi Acharya, Yuri Alexeev, K. Jordan Berg, Shouvanik Chakrabarti, Florian J. Curchod, Joan M. Dreiling, Neal Erickson, Cameron Foltz, Michael Foss- Feig, David Hayes, Travis S. Humble, Niraj...

    work page 2025

  7. [7]

    Pseudorandomness

    Salil P. Vadhan. “Pseudorandomness”. Volume 7 of Foundations and Trends in Theo- retical Computer Science. Now Publishers Inc. (2012)

    work page 2012

  8. [8]

    Extractors and pseudorandom generators

    Luca Trevisan. “Extractors and pseudorandom generators”. Journal of the ACM48, 860–879 (2001)

    work page 2001

Show all 21 references
  1. [9]

    Explicit two-source extractors and resilient functions

    Eshan Chattopadhyay and David Zuckerman. “Explicit two-source extractors and resilient functions”. Annals of Mathematics189, 653–705 (2019)

    2019

  2. [10]

    Various techniques used in connection with random digits

    John von Neumann. “Various techniques used in connection with random digits”. National Bureau of Standards Applied Mathematics Series12, 36–38 (1951)

    1951

  3. [11]

    Iterating Von Neumann’s Procedure for Extracting Random Bits

    Yuval Peres. “Iterating Von Neumann’s Procedure for Extracting Random Bits”. The Annals of Statistics20, 590–597 (1992)

    1992

  4. [12]

    From the Bernoulli factory to a dice enterprise via perfect sampling of Markov chains

    Giulio Morina, Krzysztof Łatuszyński, Piotr Nayar, and Alex Wendland. “From the Bernoulli factory to a dice enterprise via perfect sampling of Markov chains”. The Annals of Applied Probability32, 327–359 (2022)

    2022

  5. [13]

    Computational analogues of en- tropy

    Boaz Barak, Ronen Shaltiel, and Avi Wigderson. “Computational analogues of en- tropy”. In International Workshop on Randomization and Computation (RANDOM). Pages 200–215. Springer, Berlin, Heidelberg (2003)

    2003

  6. [14]

    Randomness extraction and key derivation using the cbc, cascade and hmac modes

    Yevgeniy Dodis, Krzysztof Pietrzak, and Daniel Wichs. “Randomness extraction and key derivation using the cbc, cascade and hmac modes”. In Annual International Cryp- tology Conference (CRYPTO). Pages 494–510. Springer, Berlin, Heidelberg (2004)

    2004

  7. [15]

    Free-space quantum random number generation based on spatial detection of single photons

    Bilal Khan, Faisal M. Arain, Atta ur Rehman, Kwangyun Jung, and Hyun Jae Shin. “Free-space quantum random number generation based on spatial detection of single photons”. Optics Letters 45, 4790–4793 (2020)

    2020

  8. [16]

    Quantumrandom-numbergeneration and key sharing

    J.G.Rarity, P.C.M.Owens, andP.R.Tapster. “Quantumrandom-numbergeneration and key sharing”. Journal of Modern Optics41, 2435–2444 (1994)

    1994

  9. [17]

    Tree algorithms for unbiased coin tossing with a biased coin

    Q. R. Stout and B. Warren. “Tree algorithms for unbiased coin tossing with a biased coin”. Annals of Probability12, 212–222 (1984)

    1984

  10. [18]

    Univariate discrete distributions

    Norman L. Johnson, Samuel Kotz, and Adrienne W. Kemp. “Univariate discrete distributions”. Wiley Series in Probability and Mathematical Statistics: Probability and Mathematical Statistics. Wiley. New York (1992). 2nd edition

    1992

  11. [19]

    Real time demonstration of high bitrate quan- tum random number generation with coherent laser light

    T. Symul, S. M. Assad, and P. K. Lam. “Real time demonstration of high bitrate quan- tum random number generation with coherent laser light”. Applied Physics Letters 98, 231103 (2011)

    2011

  12. [20]

    Introduction to nonparametric estimation

    Alexandre B. Tsybakov. “Introduction to nonparametric estimation”. Springer Series in Statistics. Springer. (2009). 1 edition

    2009

  13. [21]

    Detecting single infrared photons with 93% system efficiency

    F. Marsili, V. B. Verma, J. A. Stern, S. Harrington, A. E. Lita, T. Gerrits, I. Vayshenker, B. Baek, M. D. Shaw, R. P. Mirin, and S. W. Nam. “Detecting single infrared photons with 93% system efficiency”. Nature Photonics7, 210–214 (2013). 17

    2013

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.