Pith. sign in

REVIEW 3 major objections 6 minor 48 references

Finite-size Effects on The Edge Loss Probability in Non-Hermitian Quantum Walks

T0 review · 3 major / 6 minor · reviewed 2026-08-03 · deepseek-v4-flash

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

Pith's one-line read In a finite-size non-Hermitian quantum walk, boundary scattering can suppress the edge burst even when the skin effect and imaginary gap closing are present, and extreme dissipation can revive an edge burst even when the imaginary gap is op

desk verdict Finite-size extension of the edge-burst theory with a new reentrance mechanism; the numerics are credible, but the analytic prefactor and the Lyapunov-scattering link need work. read the letter →

arxiv 2512.20106 v2 pith:G5GGPXIU submitted 2025-12-23 quant-ph

Shuaixian Liu , Yulan Dong , Bowen Zeng , Mengqiu Long This is my paper · ORCID
classification quant-ph
keywords non-HermitianquantumwalkedgeburstskineffectimaginarygapLyapunovexponentboundaryscatteringfinite-sizeeffectslossprobability
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

This paper asks how the non-Hermitian edge burst—a sharp excess of loss probability at the boundary of a lossy lattice—behaves when the lattice is finite rather than infinite. The authors find that boundary scattering can suppress the burst even when the two infinite-limit ingredients, the non-Hermitian skin effect and imaginary gap closing, are both present. More surprisingly, they find that with extreme on-site dissipation, a large edge loss probability can appear even when the spectrum's imaginary gap is open, because the bulk modes effectively enter the gap-closing regime. The practical point is that in any real finite system the edge burst is controlled by the competition between bulk propagation and boundary reflection, not by the bulk spectrum alone.

What carries the argument

The central toolkit is a Green's-function residue expansion expressing the loss probability as an integral over the two roots βL/R(ω), so that the bulk loss decays with distance according to |β|^2|x−x0|. The paper combines that with two quantities: the generalized Brillouin zone radius rG = sqrt(|(t1−γ/2)/(t1+γ/2)|), which sets the strength of the skin effect, and the derivative of the Lyapunov exponent λ(v) with respect to velocity v, evaluated at v = 0, which the authors propose, by analogy with the density of states, as a measure of boundary scattering probability. The mechanism is that a large derivative means few right-moving velocities near the saddle point and therefore weak reflectio

What would settle it

Measure directly, in the same finite chain, the fraction of the wavepacket reflected from the boundary as a function of γ—for example by decomposing the time-evolved wavefunction into left- and right-moving components after the first boundary arrival—and check whether it tracks the inverse of |dλ/dv| evaluated at v = 0. If the reflected fraction does not fall as |dλ/dv| rises, the scattering proxy is falsified.

Watch

Extended reading notes

Core claim

On the paper's terms, the discovery is that the finite-size edge loss probability P1 as a function of dissipation γ separates into regimes. For moderate γ, a strong skin effect (small generalized Brillouin zone radius) makes boundary scattering weak, so the wavepacket is trapped at the edge and the power-law bulk loss from imaginary gap closing produces P1 ≫ Pmin. For very small or very large γ, boundary scattering becomes strong, and even a properly closed imaginary gap fails to produce an edge burst. Independently, when |t1| > |t2| so the periodic-boundary spectrum has an imaginary gap, raising γ to extreme values pushes min|βL(ω)| back toward 1, so some bulk modes behave as if the gap wer

Load-bearing premise

The analysis treats the derivative of the Lyapunov exponent at zero velocity, |dλ/dv| at v = 0, as the probability of boundary scattering by analogy with the density of states, without a derivation; if that identification fails, the proposed mechanism for suppression and reemergence would need revision, although the numerical findings themselves could still stand.

Editorial extensions

If this is right

  • In finite-size chains, the infinite-limit conditions for an edge burst (NHSE plus imaginary gap closing) are necessary but not sufficient; boundary reflection can erase the burst.
  • At extreme dissipation, the effective closing of the imaginary gap for bulk modes can restore the power-law bulk decay, making an edge burst possible even though the periodic-boundary spectrum retains an imaginary gap.
  • Analytical estimates of the edge loss probability from the infinite-limit residue formula agree with numerics only in the weak-scattering parameter window; outside it, boundary-scattering corrections are required.
  • The direction of the edge burst is set by the group velocity v = −t1 of the dominant leftward mode, and the strongest boundary trapping occurs at the non-Bloch PT transition rG = 0, where the open-boundary spectrum collapses to a point.

Reading between the lines

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

  • If the Lyapunov-derivative proxy holds, a general finite-size rule may be that an edge burst appears whenever the saddle-point velocity distribution is narrow enough that the boundary reflects little of the packet; this rule could be tested in other non-reciprocal lattices with tunable loss.
  • A testable experimental prediction is that in photonic or cold-atom quantum walks the edge loss probability should be non-monotonic in dissipation, with suppression and revival peaks tracking where |dλ/dv| at v = 0 and the skin-effect strength vary, rather than simply where the imaginary gap closes.
  • The paper implicitly extends the dynamical bulk-edge correspondence to finite systems: in finite lattices the relevant parameter is not only the bulk spectrum but also a boundary scattering kernel, so a phase diagram in the (t1, γ) plane separating burst, suppression, and reemergence would be a natural next step.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. The paper studies finite-size effects on the non-Hermitian edge burst in a two-chain lossy quantum-walk model previously treated in the thermodynamic limit by Xue et al. [31]. For a finite chain with L=100 and an edge at x=1, the authors numerically compute the edge loss probability P1 as a function of the dissipation γ, for parameter sets that in the infinite limit satisfy either imaginary-gap closing or imaginary-gap opening of the PBC spectrum. They find that boundary scattering can suppress the edge burst even when the infinite-limit conditions are met, and, conversely, that at extreme dissipation an edge burst can reappear even when the PBC imaginary gap is open. The authors propose an explanation based on the competition between bulk propagation and boundary scattering, quantified by the GBZ radius and by the derivative of the Lyapunov exponent at v=0.

Significance. If correct, the paper gives a useful finite-size refinement of the dynamical bulk-edge correspondence: the edge burst is not solely a bulk-spectrum effect, and boundary reflection must be included. The numerical simulations are direct and parameter-free; no fitted quantities are introduced. The comparison with the infinite-chain benchmark is honest, and the regimes where the analytic approximation fails are explicitly acknowledged. The main weakness is that the central mechanistic ingredient—that |dλ(v)/dv| at v=0 controls the boundary-scattering probability—is asserted by analogy rather than derived or independently validated. The quantitative analytic anchor also contains a prefactor inconsistency. These issues are fixable and do not invalidate the numerical observations, but they do need to be addressed before the explanation can be regarded as established.

major comments (3)
  1. [Sec. III, Eqs. (7), (10)] Direct evaluation of Eq. (7) with the stated n=m=2 expansions |f_L|^2≈Q δω^2 and |β_L|≈1+K δω^2 gives P_x = γ/(2√π) Q(2K)^{-3/2}|x-x0|^{-3/2}, not the stated 2γΓ(3/2)/π Q(2K)^{-3/2}|x-x0|^{-3/2} = γ/√π Q(2K)^{-3/2}|x-x0|^{-3/2}. The factor of 2 propagates into Eqs. (11)–(13). The qualitative γ-dependence in Eq. (13) is unchanged, but the quantitative comparison in Fig. 2(a) and the analytic benchmark need to be corrected, or explicitly attributed to Ref. [31] if that is the intended source.
  2. [Sec. III, Eq. (11)] Equation (11) is obtained by integrating the infinite-line bulk formula (10) over x=-∞…0. This is not a model of a finite boundary at x=1; it simply sums an asymptotic bulk solution and cannot by itself capture trapping or reflection at the physical edge. The authors note that Eq. (11) gives unphysical P1>1 for small γ and for sufficiently large γ—precisely the regimes where suppression of the edge burst is claimed. Thus the analytic curve does not independently support the central suppression claim; the evidence for suppression is the numerical P1(γ), and the mechanism must be justified separately.
  3. [Sec. III, Fig. 3(d), text after Eq. (14)] The assertion that |dλ(v)/dv|_{v=0} quantifies the boundary-scattering probability is introduced only 'by analogy with the density of states' and is not derived from the microscopic dynamics or checked against a reflected-current/amplitude diagnostic. The use of the absolute value is particularly problematic: a positive derivative means rightward-moving velocities have larger λ than v=0 and would enhance scattering, whereas a negative derivative would suppress it. Unless dλ/dv<0 for all relevant γ, |dλ/dv| conflates opposite mechanisms. The manuscript needs either a derivation of the scattering probability from a saddle-point/reflection calculation or an independent numerical validation, and the signed derivative should be used.
minor comments (6)
  1. [Eq. (5)] The integration variable is written as 'dw' but should be 'dω'.
  2. [Throughout] Several typos: 'dominat' after Eq. (8), 'mights expect' before Eq. (10), 'polying' in Sec. III, and a Chinese comma in 'propagation，as shown'.
  3. [Sec. IV] The phrase 'transition from the condition of imaginary gap opening to imaginary gap closing' is stronger than what is shown: min[|β_L(ω)|] approaching 1 means the modes approach the real axis, but the PBC imaginary gap likely remains finite. Please quantify or soften this claim.
  4. [Fig. 3(b)] The red lines marking ω0 on the PBC spectrum are mentioned but are not legible in the figure; please enlarge or annotate them.
  5. [Fig. 2(a)] The paper defines edge burst as P1≫Pmin, but no quantitative threshold is given. A concrete criterion would help distinguish 'sizable' from 'burst' in Figs. 1 and 4.
  6. [Availability] No data/code availability statement is included. For a numerical study, a reproducibility statement or a statement of data availability would be helpful.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: finite-size P1 is computed from model parameters, and the heuristic Lyapunov-derivative diagnostic is an asserted correlator rather than an input to the calculation.

full rationale

The derivation is self-contained. The finite-size edge loss probability is computed directly by evolving Eq. (1) and integrating Eq. (3), while the analytical estimates (Eqs. (10)-(11) and the Sec. IV expansions) are obtained from the model parameters t1, t2, and γ via the Green's function and residue expansions, with no parameters fitted to the numerical P1 curves. The infinite-limit edge-burst criterion of Xue et al. [31] is used as an external benchmark, not as an input to the finite-size calculation. The only non-derived element is the scattering diagnostic |dλ/dv| at v=0, introduced 'by analogy with the density of states' (Sec. III, Fig. 3(d)); this is a heuristic explanatory correlator, not a fitted parameter and not a redefinition of P1, so the paper's main numerical findings do not logically reduce to it. The self-citation [48] for the DOS analogy is not load-bearing evidence for a prediction and does not create a circular derivation chain. No self-definitional, fitted-input-called-prediction, or self-citation-chain circularity is present.

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

No free parameters are fitted: t1, t2, γ, L, x0 are model inputs and Q, K are derived from those inputs. No new physical entities are introduced. The main nontrivial axioms are the standard GBZ/Lyapunov machinery and the ad hoc identification of dλ/dv|v=0 with scattering probability.

assumptions (4)
  • domain assumption The dynamics is governed by a non-Hermitian Schrödinger equation and the norm decrease is interpreted as particle loss (Eqs. (1)-(3)).
    Standard for lossy quantum walks; the loss rate is taken as 2γ|ψ_B|².
  • domain assumption The Green's function representation and residue calculus for an infinite chain (Eqs. (5)-(7)) provide the baseline edge-loss formula.
    Uses standard non-Hermitian Green's function methods and GBZ theory [13,23,24].
  • ad hoc to paper Boundary scattering probability is proportional to the reciprocal of |dλ(v)/dv| at v=0, by analogy with density of states (Fig. 3(d), Sec. III).
    Asserted via analogy to Ref. [48]; no derivation is given, but it is used to explain the parameter dependence of scattering.
  • domain assumption The saddle-point/Lefschetz-thimble evaluation provides the dominant Lyapunov exponent for wavepacket propagation (Eq. (14) and Fig. 3).
    Assumes validity of complex-saddle analysis for non-Hermitian wavepacket dynamics [44].

how reviews work

0 comments
Cite this review

Pith. "Pith review of Finite-size Effects on The Edge Loss Probability in Non-Hermitian Quantum Walks." pith.science (2026). https://pith.science/paper/G5GGPXIU

@misc{pith2026251220106,
  author       = {Pith},
  title        = {Pith review of: Finite-size Effects on The Edge Loss Probability in Non-Hermitian Quantum Walks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/G5GGPXIU}},
  note         = {Machine review of arXiv:2512.20106}
}
read the original abstract

A dynamical bulk-edge relation in quantum walks has been theoretically proposed and experimentally observed, in which a power-law dependence of the bulk loss probability is associated with a pronounced peak of loss probability at the edge. This behavior has been proven to arise from imaginary gap closing and the non-Hermitian skin effect in the infinite limit without boundary effects. However, in a finite-size chain, we find that boundary scattering can suppress this edge burst. Meanwhile, imaginary gap opening, together with the non-Hermitian skin effect, can also induce a large loss probability at the edge. Our results provide insights into finite-size quantum dynamics.

Figures

Figures reproduced from arXiv: 2512.20106 by the authors.

Figure 1
Figure 1. FIG. 1. (a) Coupled two chains with [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 3
Figure 3. FIG. 3. Lyapunov exponent for (a) negative velocity for dif [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. FIG. 4. (a) Under the condition of imaginary gap opening, [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figures from the paper (1 more)
Figure 5
Figure 5. Figure 5: FIG. 5. Scale of bulk loss probability with respect to relative [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

48 extracted references · 1 linked inside Pith

  1. [31]

    Xue, Y.-M

    W.-T. Xue, Y.-M. Hu, F. Song, and Z. Wang, Physical Review Letters128, 120401 (2022)

    2022

  2. [1]

    W.-P. Su, J. Schrieffer, and A. Heeger, Phys. Rev. B22, 2099 (1980)

  3. [2]

    M. V. Berry, Proc. R. Soc. London, Ser. A392, 45 (1984)

    1984

  4. [3]

    A. J. Heeger, S. Kivelson, J. Schrieffer, and W.-P. Su, Rev. Mod. Phys.60, 781 (1988)

    1988

  5. [4]

    Zak, Phys

    J. Zak, Phys. Rev. Lett.62, 2747 (1989)

    1989

  6. [5]

    M. Z. Hasan and C. L. Kane, Rev. Mod. Phys.82, 3045 (2010)

    2010

  7. [6]

    Qi and S.-C

    X.-L. Qi and S.-C. Zhang, Rev. Mod. Phys.83, 1057 (2011)

    2011

  8. [7]

    B. A. Bernevig, inTopological Insulators and Topological Superconductors(Princeton university press, 2013). 7

    2013

Show all 48 references
  1. [8]

    J. K. Asb´ oth, L. Oroszl´ any, and A. P´ alyi, Lecture notes in physics919, 166 (2016)

    2016

  2. [9]

    Hatano and D

    N. Hatano and D. R. Nelson, Phys. Rev. Lett.77, 570 (1996)

    1996

  3. [10]

    T. E. Lee, Phys. Rev. Lett.116, 133903 (2016)

    2016

  4. [11]

    Xiong, J

    Y. Xiong, J. Phys. Commun.2, 035043 (2018)

    2018

  5. [12]

    F. K. Kunst, E. Edvardsson, J. C. Budich, and E. J. Bergholtz, Phys. Rev. Lett.121, 026808 (2018)

    2018

  6. [13]

    Yao and Z

    S. Yao and Z. Wang, Physical review letters121, 086803 (2018)

    2018

  7. [14]

    K. Zhou, Z. Yang, B. Zeng, and Y. Hu, Science China Physics, Mechanics & Astronomy69, 217211 (2026)

    2026

  8. [15]

    Longhi, Physical Review Research1, 023013 (2019)

    S. Longhi, Physical Review Research1, 023013 (2019)

    2019

  9. [16]

    Kawabata, K

    K. Kawabata, K. Shiozaki, M. Ueda, and M. Sato, Phys. Rev. X9, 041015 (2019)

    2019

  10. [17]

    Okuma, K

    N. Okuma, K. Kawabata, K. Shiozaki, and M. Sato, Phys. Rev. Lett.124, 086801 (2020)

    2020

  11. [18]

    Ashida, Z

    Y. Ashida, Z. Gong, and M. Ueda, Adv. Phys.69, 249 (2020)

    2020

  12. [19]

    Zhang, T

    X. Zhang, T. Zhang, M.-H. Lu, and Y.-F. Chen, Ad- vances in Physics: X7, 2109431 (2022)

    2022

  13. [20]

    R. Lin, T. Tai, L. Li, and C. H. Lee, Frontiers of Physics 18, 53605 (2023)

    2023

  14. [21]

    H.-Y. Wang, F. Song, and Z. Wang, Physical Review X 14, 021011 (2024)

    2024

  15. [22]

    Zeng and T

    B. Zeng and T. Yu, Physical Review Research5, 013003 (2023)

    2023

  16. [23]

    Yokomizo and S

    K. Yokomizo and S. Murakami, Physical review letters 123, 066404 (2019)

    2019

  17. [24]

    Z. Yang, K. Zhang, C. Fang, and J. Hu, Physical Review Letters125, 226402 (2020)

    2020

  18. [25]

    K. Ding, C. Fang, and G. Ma, Nature Reviews Physics , 1 (2022)

    2022

  19. [26]

    Zhang, Z

    K. Zhang, Z. Yang, and C. Fang, Physical Review Letters 125, 126402 (2020)

    2020

  20. [27]

    E. J. Bergholtz, J. C. Budich, and F. K. Kunst, Rev. Mod. Phys.93, 015005 (2021)

    2021

  21. [28]

    Okuma and M

    N. Okuma and M. Sato, Annual Review of Condensed Matter Physics14, 83 (2023)

    2023

  22. [29]

    T. Yu, J. Zou, B. Zeng, J. Rao, and K. Xia, Physics Reports1062, 1 (2024)

    2024

  23. [30]

    Z. Yang, K. Zhou, B. Zeng, and Y. Hu, Physical Review B111, L041406 (2025)

    2025

  24. [32]

    L. Wang, Q. Liu, and Y. Zhang, Chinese Physics B30, 020506 (2021)

    2021

  25. [33]

    Hu, W.-T

    Y.-M. Hu, W.-T. Xue, F. Song, and Z. Wang, Physical Review B108, 235422 (2023)

    2023

  26. [34]

    Ren and X.-G

    C.-C. Ren and X.-G. Yin, ACTA PHYSICA SINICA72 (2023)

    2023

  27. [35]

    Xiao, W.-T

    L. Xiao, W.-T. Xue, F. Song, Y.-M. Hu, W. Yi, Z. Wang, and P. Xue, Physical Review Letters133, 070801 (2024)

    2024

  28. [36]

    Zhu, Y.-L

    J. Zhu, Y.-L. Mao, H. Chen, K.-X. Yang, L. Li, B. Yang, Z.-D. Li, and J. Fan, Physical Review Letters132, 203801 (2024)

    2024

  29. [37]

    P. Wen, J. Pi, and G.-L. Long, Physical Review A109, 022236 (2024)

    2024

  30. [38]

    Yuce and H

    C. Yuce and H. Ramezani, Physical Review B109, 214301 (2024)

    2024

  31. [39]

    Sen and C

    G. Sen and C. Yuce, Physical Review B111, 174305 (2025)

    2025

  32. [40]

    S. Ma, H. Lin, and J. Pi, Physical Review B109, 214311 (2024)

    2024

  33. [41]

    Hu, H.-Y

    Y.-M. Hu, H.-Y. Wang, Z. Wang, and F. Song, Physical Review Letters132, 050402 (2024)

    2024

  34. [42]

    Xiong and H

    Y. Xiong and H. Hu, Physical Review B109, L100301 (2024)

    2024

  35. [43]

    W.-T. Xue, F. Song, Y.-M. Hu, and Z. Wang, arXiv preprint arXiv:2503.13671 (2025)

    2025 arXiv

  36. [44]

    Yang and C

    T.-H. Yang and C. Fang, Physical Review Letters135, 186401 (2025)

    2025

  37. [45]

    Witten, AMS/IP Stud

    E. Witten, AMS/IP Stud. Adv. Math50, 347 (2011)

    2011

  38. [46]

    Mukherjee and M

    A. Mukherjee and M. Cristoforetti, Phys. Rev. B90, 035134 (2014)

    2014

  39. [47]

    Kanazawa and Y

    T. Kanazawa and Y. Tanizaki, Journal of High Energy Physics2015, 1 (2015)

    2015

  40. [48]

    Zeng, Z.-K

    B. Zeng, Z.-K. Ding, H. Pan, N. Luo, J. Zeng, L.-M. Tang, and K.-Q. Chen, Applied Physics Letters121, 252202 (2022)

    2022

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.