Pith. sign in

REVIEW 2 major objections 5 minor 26 references

Semi-Discrete Linear Hyperbolic Curvature flow

T0 review · 2 major / 5 minor · reviewed 2026-08-27 · deepseek-v4-flash

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

Pith's one-line read This paper claims that the semi-discrete hyperbolic curvature flow with prescribed endpoints is completely solvable: every self-similar motion falls into one of four classes, and for arbitrary boundary trajectories the unique solution is…

desk verdict Competent extension with explicit solution formulas and a correct self-similar classification, but the main diagonal formula fails at critical damping where the system matrix becomes defective. read the letter →

arxiv 2608.18826 v1 pith:QP4OICKQ submitted 2026-08-19 math.DS math.CA

James McCoy , Rohan St Hill This is my paper · ORCID
classification math.DSmath.CA MSC 34A2637C60
keywords semi-discretecurvatureflowhyperbolicgeometricself-similarsolutionsquadraticeigenvalueproblempiecewiselinearcurvesdampedwaveequationcurveshorteningdifference
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 studies a discrete model of inertia in curvature-driven motion: each interior vertex of a piecewise linear curve accelerates in the direction of its discrete curvature, with damping proportional to velocity, while the two endpoints move along prescribed trajectories. The central claim is that this second-order geometric evolution is completely solvable. All self-similar evolutions — scaling, rotation, translation, their combinations, and the static straight-line interpolant — are classified, and for any initial curve and any endpoint motion the unique solution is written down explicitly using the eigenmodes of the discrete Laplacian. The same machinery produces a curvature-difference flow that evolves any piecewise linear curve to any other with the same number of vertices. A sympathetic reader would care because this makes a nonlinear-looking geometric flow exactly solvable, exposing the same spectral structure that governs damped vibrating chains.

What carries the argument

The central object is the block matrix $M = \begin{pmatrix} 0 & I \\ A & -\beta I \end{pmatrix}$ obtained by reducing the second-order flow to a first-order system; its eigenvalues solve the quadratic eigenvalue problem $\lambda^2 + \beta\lambda - \mu_k = 0$, where $\mu_k = -2 + 2\cos(k\pi/(n-1))$ are the eigenvalues of the discrete Laplacian $A = \operatorname{tridiag}(1,-2,1)$ with Dirichlet boundary conditions. This spectral data produces the matrix exponential $e^{Mt}$, whose diagonalised form $P \operatorname{diag}(e^{\lambda_j t}) P^{-1}$ carries the variation-of-parameters solution (16) and the closed-form constant-forcing version (18). The self-similar classification instead runs through the Sylvester equation $A\mathbf{U}_0 - \mathbf{U}_0 M_0 = -\alpha$, which decides when a prescribed scaling-rotation admits a unique interior profile. The discrete energy $E = K + W$ and its dissipation law $dE/dt = -2\beta K$ control long-time behaviour, including convergence to the straight-line interpolant.

What would settle it

Choose $n = 5$ and set $\beta = 2\sqrt{2 - 2\cos(\pi/4)} \approx 1.531$, so that the discriminant $\beta^2 + 4\mu_1$ vanishes for the first mode. Compute the null space of $M - (-\beta/2)I$; if it has dimension one while the eigenvalue has algebraic multiplicity two, the matrix is not diagonalisable and formula (16) is undefined at a value of $\beta$ the paper claims to cover.

Watch

Extended reading notes

Core claim

The load-bearing discovery is Theorem 2 combined with Theorem 1. Theorem 2 states that the initial-value problem for the semi-discrete hyperbolic curvature flow with arbitrary boundary trajectories has a unique solution, given by the variation-of-parameters formula (16), in which the first-order system matrix $M = \begin{pmatrix} 0 & I \\ A & -\beta I \end{pmatrix}$ is diagonalised through the eigenvalues of the tridiagonal Toeplitz matrix $A = \operatorname{tridiag}(1,-2,1)$. Theorem 1 states that every self-similar solution, for a non-degenerate initial curve and self-similar boundary data in the plane, falls into exactly one of four mutually exclusive classes: pure scaling-rotation, pure translation, scaling-rotation with translation, or static equilibrium. In the scaling-rotation classes, the interior profile is the unique solution of a Sylvester equation $A\mathbf{U}_0 - \mathbf{U}_0 M_0 = -\alpha$, with existence controlled by disjointness of the spectra of $A$ and $M_0$. The paper also proves a hyperbolic curvature-difference flow that carries any piecewise linear curve to any other with the same vertex count, with exponential convergence for $\beta > 0$.

Load-bearing premise

The whole argument leans on the assumption that the system matrix can always be diagonalised; at special damping values two modes merge and only one eigenvector remains, so that assumption fails and the formula's eigenvector matrix does not exist.

Editorial extensions

If this is right

  • All self-similar evolutions of the flow fall into exactly four classes — pure scaling-rotation, pure translation, combined scaling-rotation-translation, and static equilibrium — with no other shapes possible.
  • For any initial curve and any prescribed endpoint trajectories there is a unique solution, given explicitly by a finite spectral sum; with constant boundary data the integral term is evaluated in closed form.
  • With fixed boundary points, every solution converges to the straight line interpolant between the endpoints, and the discrete energy decreases monotonically at rate $-2\beta K$.
  • For $\beta > 0$ and $T$-periodic boundary forcing, a unique $T$-periodic solution exists and is globally attracting.
  • The hyperbolic curvature-difference flow evolves any piecewise linear curve to any other with the same number of vertices, converging exponentially when the damping is positive.

Reading between the lines

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

  • Beyond the paper's own claims, the critical-damping values where $\beta^2 + 4\mu_k = 0$ fall inside the stated domain $\beta \ge 0$; a Jordan-form correction, adding factors like $t e^{-\beta t/2}$, would likely extend the explicit solution formula to those parameter values.
  • Because the coefficient matrix is the path graph Laplacian, the open-curve results suggest that wave-diffusion processes on graphs with Dirichlet boundaries are explicitly solvable, and that periodic boundary forcing creates a boundary-driven resonance that has no analogue on closed graphs.
  • The coordinate-wise anisotropic variant decouples by ambient coordinate, so the predicted evolutions could be checked directly with a physical mass-spring chain whose endpoint displacements follow the prescribed trajectories.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Request a human review

A listed scientist reviews the paper for a fee and the review publishes here regardless of verdict. See the reviewers or get listed.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 5 minor

Summary. The paper studies the semi-discrete hyperbolic curvature flow (SDHF) of piecewise linear curves with prescribed boundary trajectories, defined by U¨ + βU˙ = A U + f(t). The authors derive an energy dissipation law, compute the spectra of the spatial matrix A and the first-order block matrix M, classify all self-similar solutions into four classes (pure scaling-rotation, pure translation, scaling-rotation with translation, and static equilibrium), construct time-periodic solutions, give an explicit variation-of-parameters solution formula for general initial data and arbitrary boundary forcing, introduce a hyperbolic curvature difference flow that evolves one polygonal curve to another, and present anisotropic and matrix-coefficient variants. The main theoretical tools are the quadratic eigenvalue problem, Sylvester's theorem, and the matrix exponential.

Significance. If the claims are correct, the paper provides a complete explicit theory of the linear semi-discrete hyperbolic flow with boundary: a closed-form solution for all times and boundary data, a full classification of self-similar solutions (including translations, which are impossible in the closed-polygon setting), and a Yau-type flow between arbitrary polygonal curves. The reduction of the self-similar classification to a Sylvester equation and the treatment of the damped quadratic eigenvalue problem are elegant and likely useful for the wider discrete geometric flow community. The paper is largely self-contained and its spectral formulas are explicit and directly verifiable. However, the main representation theorem is invalid at critical damping values where the first-order matrix is defective, and one anisotropic solution formula is incorrect; these issues are local and repairable but currently limit the advertised generality.

major comments (2)
  1. [§3.3, Theorem 2 (with Proposition 2)] The proof of Theorem 2 asserts that the matrix M = [[0,I],[A,-βI]] is diagonalisable for every β ≥ 0, with eigenvector matrix P from Proposition 2. This fails at critical damping: when β² + 4µ_k = 0 for some mode k, the two quadratic roots coincide at λ = -β/2 and Proposition 2 supplies two identical eigenvectors (u_k, -β/2 u_k)^⊤ for that mode, so P is singular and the displayed diagonal formula (16) is undefined. The exceptional values are nonempty (e.g., n=4, m=2, k=1 gives β=2). The defect propagates to Corollary 3, Theorem 3, Corollary 4, and Proposition 11, all of which invoke P diag(...)P^{-1}. Existence and uniqueness of the linear ODE do survive through the matrix exponential, and a Jordan-form correction or an explicit restriction on β would repair the statement; as written, the theorem overclaims validity for all β ≥ 0.
  2. [§5.1, Proposition 9] The solution formula for the coordinate-wise anisotropic flow (23) uses the eigenvector matrix P of the block matrix for the isotropic operator A and the same eigenvalues λ_i, rescaled by c_j: P diag(e^{c_j λ_i t})P^{-1}. But for the coordinate equation with coefficient matrix c_j A, the relevant eigenvalues are the roots of λ² + βλ - c_j µ_k = 0, not c_j λ_i, and the corresponding eigenvector matrix P_j has columns (u_k, λ u_k) with these new λ, so P_j ≠ P unless c_j = 1. A simple scalar check (m=1, β=0, c_j=2: the true eigenvalues are ±2, while c_j λ_i gives ±2√2) confirms the formula is incorrect. The proof ("follows in the same manner as Theorem 2") does not justify the rescaling, and the displayed solution should be corrected or the proposition removed.
minor comments (5)
  1. [§2.1, Eq. (3)] The discrete Dirichlet energy W sums i=2 to n-1, omitting the edge (X_{n-1},X_n). With this definition the summation-by-parts identity in the proof leaves the boundary term <X_n - X_{n-1}, dX_{n-1}/dt>, so the claimed identity (5) and the dissipation law (4) do not follow. The correct statement holds if the sum is taken to i=n; please correct the index in (3).
  2. [§3.1, Theorem 1] The four classes are not mutually exclusive as stated, since the static equilibrium (g≡1, R≡I, h≡0) is a special case of the pure scaling-rotation class (i) with γ=0, ω0=0 and satisfies the Sylvester equation (13) with M0=0. Recommend phrasing (i) as the non-trivial scaling-rotation case or explicitly noting the overlap.
  3. [§3.3, Corollary 3 proof] In the displayed evaluation of the integral, the terms (e^{λ_i} - 1)/λ_i are missing the time variable; they should read (e^{λ_i t} - 1)/λ_i.
  4. [§4, Theorem 3 statement] The condition "fn = Xn(0)" should be "fn(0) = Xn(0)" to be consistent with the boundary condition notation.
  5. [§2.2, Proposition 2] The index "j=1,2,...,2m" in the eigenvalue listing is misleading; for each k there are exactly two eigenvalues, so the set should be written as λ_k^±.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the central solution and classification theorems are derived from the flow definition and standard ODE results, with self-citations used only as supporting context or lemmas.

full rationale

The paper's main derivation chain is self-contained. The flow (SDHF) is defined independently as a damped second-order linear system, and the solution formula (Theorem 2), the self-similar classification (Theorem 1), and the difference-flow construction (Theorem 3) follow by substituting this definition into standard elementary facts: the spectral decomposition of a tridiagonal Toeplitz matrix (Lemma 1), the quadratic eigenvalue problem for the first-order block matrix (Proposition 2), variation of parameters, and the classical Sylvester solvability condition. No fitted parameter is renamed as a prediction, and no conclusion is used as one of its own hypotheses. The self-similar ansatz (9) is explicitly imposed in Definition 3, so classifying its solutions into scaling-rotation, translation, or combinations is the intended analysis rather than a hidden assumption. Self-citations appear mainly as context: [10] is cited for closed-polygon breather behaviour, and [11] is used once in Proposition 10(i) for an eigenvalue property of a vertex-weighted Laplacian, a parameter-free supporting lemma from a prior published article that does not assume the hyperbolic claims made here. This is a routine citation of prior work, not a load-bearing circularity. The noted critical-damping obstruction to the diagonal formula in Theorem 2 is a correctness or domain-of-validity issue, not a circularity: the formula is invalid at certain parameter values rather than equivalent to its inputs. Overall, no circular step reduces a derived result to the data from which it was obtained.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

No entities are invented. The free parameters listed (β and the anisotropy constants) are model constants, not fitted to data. The axiomatic content is standard linear algebra and linear ODE theory, plus the modeling choice of the discrete flow itself. One self-cited eigenvalue result (Proposition 10(i)) is a published theorem from the authors' previous paper, not a circular input.

free parameters (3)
  • β (damping coefficient)
    Introduced in Definition 2 as β ≥ 0; the central solution formulas and classifications are parameterised by β but hold for all values aside from the non-diagonalisable resonance. It is a model parameter, not fitted.
  • c_j (coordinate-wise anisotropy)
    Positive constants in Definition 5; arbitrary model inputs, not fitted.
  • c_i (vertex-wise anisotropy)
    Positive mobility constants in Definition 6; arbitrary model inputs, not fitted.
assumptions (6)
  • standard math Eigenvalues and eigenvectors of the tridiagonal Toeplitz matrix tridiag(c,a,b) are as stated in Lemma 1.
    Used throughout to diagonalise A and M; cited to Smith [17].
  • standard math The Sylvester equation AX - XB = C has a unique solution iff the spectra of A and B are disjoint.
    Invoked in Proposition 5; cited to Horn and Johnson [23].
  • standard math Variation of parameters formula for linear ODE systems with matrix exponential.
    Basis for Theorems 2, 3, and Proposition 11; cited to Hartman [24].
  • domain assumption Discrete curvature at an interior vertex is the second difference Xi-1 - 2Xi + Xi+1.
    Definition 1; this is the standard semi-discrete analogue used in [1, 11].
  • domain assumption The interior vertices accelerate in the direction of discrete curvature with damping: d2Xi/dt2 + β dXi/dt = Xi-1 - 2Xi + Xi+1.
    Definition 2, motivated in Section 2.1 as the centred finite difference discretisation of the damped wave equation.
  • standard math In the vertex-wise anisotropic flow, T = diag(c_i)A has real, negative, simple eigenvalues.
    Proposition 10(i), quoted from the authors' prior paper [11, Proposition 5.2].

how reviews work

0 comments
Cite this review

Pith. "Pith review of Semi-Discrete Linear Hyperbolic Curvature flow." pith.science (2026). https://pith.science/paper/QP4OICKQ

@misc{pith2026260818826,
  author       = {Pith},
  title        = {Pith review of: Semi-Discrete Linear Hyperbolic Curvature flow},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QP4OICKQ}},
  note         = {Machine review of arXiv:2608.18826}
}
read the original abstract

We consider evolution of piecewise linear curves with boundary by linear hyperbolic curvature flows. Given appropriate boundary behaviours we find corresponding self-similar solutions. More generally, given any specific boundary trajectories we write down general solutions using finite Fourier expansions. Finally we consider the problem of evolving one piecewise linear curve to another using nonhomogeneous linear hyperbolic flow.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

26 extracted references · 26 canonical work pages

  1. [1]

    The American Mathematical Monthly 114(4), 316–328 (2007)

    Chow, B., Glickenstein, D.: Semidiscrete geometric flows of polygons. The American Mathematical Monthly 114(4), 316–328 (2007)

    work page 2007

  2. [2]

    Results in Mathematics 71(1-2), 455–482 (2016) https://doi.org/10.1007/ s00025-016-0572-5

    Rademacher, C., Rademacher, H.-B.: Solitons of Discrete Curve Shorten- ing. Results in Mathematics 71(1-2), 455–482 (2016) https://doi.org/10.1007/ s00025-016-0572-5

    work page 2016

  3. [3]

    The Jour- nal of Geometric Analysis 28(3), 2902–2925 (2017) https://doi.org/10.1007/ s12220-017-9940-y

    Glickenstein, D., Liang, J.: Asymptotic Behavior of β-Polygon Flows. The Jour- nal of Geometric Analysis 28(3), 2902–2925 (2017) https://doi.org/10.1007/ s12220-017-9940-y

    work page 2017

  4. [4]

    The Journal of Geometric Analysis 35(11) (2025) https://doi.org/10.1007/ s12220-025-02195-x 34

    McCoy, J., Meyer, J.: Linear Semi-Discrete Polyharmonic Flows of Closed Poly- gons. The Journal of Geometric Analysis 35(11) (2025) https://doi.org/10.1007/ s12220-025-02195-x 34

    work page 2025

  5. [5]

    SIAM Journal on Mathematical Analysis 22(3), 575–586 (1991) https://doi.org/10.1137/0522036

    Gurtin, M.E., Podio-Guidugli, P.: A Hyperbolic Theory for the Evolution of Plane Curves. SIAM Journal on Mathematical Analysis 22(3), 575–586 (1991) https://doi.org/10.1137/0522036

    work page doi:10.1137/0522036 1991

  6. [6]

    Journal of Crystal Growth 198-199, 1256–1261 (1999) https://doi.org/10

    Rotstein, H.G., Brandon, S., Novick-Cohen, A.: Hyperbolic flow by mean curva- ture. Journal of Crystal Growth 198-199, 1256–1261 (1999) https://doi.org/10. 1016/s0022-0248(98)01006-9

    work page 1999

  7. [7]

    The Journal of Chemical Physics133(17), 174106 (Nov 2010).https://doi.org/10.1063/1

    Kong, D.-X., Liu, K.: Wave character of metrics and hyperbolic geometric flow. Journal of Mathematical Physics 48(10) (2007) https://doi.org/10.1063/1. 2795839

    work page doi:10.1063/1 2007

  8. [8]

    Journal of Dif- ferential Equations 246(1), 373–390 (2009) https://doi.org/10.1016/j.jde.2008

    He, C.-L., Kong, D.-X., Liu, K.: Hyperbolic mean curvature flow. Journal of Dif- ferential Equations 246(1), 373–390 (2009) https://doi.org/10.1016/j.jde.2008. 06.026

    work page doi:10.1016/j.jde.2008 2009

Show all 26 references
  1. [9]

    Journal de Mathématiques Pures et Appliquées 90(6), 591–614 (2008) https://doi.org/10

    LeFloch, P.G., Smoczyk, K.: The hyperbolic mean curvature flow. Journal de Mathématiques Pures et Appliquées 90(6), 591–614 (2008) https://doi.org/10. 1016/j.matpur.2008.09.006

    2008

  2. [10]

    Mathematics in Engineering 7(3), 281–315 (2025) https://doi.org/10

    McCoy, J., Meyer, J.: Semi-discrete linear hyperbolic polyharmonic flows of closed polygons. Mathematics in Engineering 7(3), 281–315 (2025) https://doi.org/10. 3934/mine.2025013

    2025

  3. [11]

    Geometriae Dedicata 220(3) (2026) https://doi.org/10.1007/ s10711-026-01080-3

    McCoy, J., St Hill, R.: Semi-discrete linear curvature flows of curves with boundary. Geometriae Dedicata 220(3) (2026) https://doi.org/10.1007/ s10711-026-01080-3

    2026

  4. [12]

    SIAM Review 43(2), 235–286 (2001)

    Tisseur, F., Meerbergen, K.: The quadratic eigenvalue problem. SIAM Review 43(2), 235–286 (2001)

    2001

  5. [13]

    https://doi.org/10.1090/cbms/092

    Chung, F.: Spectral Graph Theory (1996). https://doi.org/10.1090/cbms/092 . http://dx.doi.org/10.1090/cbms/092

    1996 doi

  6. [14]

    IEEE Signal Processing Magazine 30(3), 83–98 (2013) https://doi.org/10.1109/msp.2012.2235192

    Shuman, D.I., Narang, S.K., Frossard, P., Ortega, A., Vandergheynst, P.: The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains. IEEE Signal Processing Magazine 30(3), 83–98 (2013) https://doi.org/10....

    2013

  7. [15]

    In: Meila, M., Zhang, T

    Chamberlain, B., Rowbottom, J., Gorinova, M.I., Bronstein, M., Webb, S., Rossi, E.: Grand: Graph neural diffusion. In: Meila, M., Zhang, T. (eds.) Proceed- ings of the 38th International Conference on Machine Learning. Proceedings of Machine Learning Research, vol. 139, pp. 14...

    2021

  8. [16]

    In: Proceedings of the ACM on Web Conference 35 2025, pp

    Yue, J., Li, H., Sheng, J., Guo, Y., Zhang, X., Zhou, C., Liu, T., Guo, L.: Graph Wave Networks. In: Proceedings of the ACM on Web Conference 35 2025, pp. 1365–1379 (2025). https://doi.org/10.1145/3696410.3714673 . ACM. http://dx.doi.org/10.1145/3696410.3714673

    2025

  9. [17]

    Oxford Press, ??? (1979)

    Smith, G.D.: Numerical Solution of Partial Differential Equations. Oxford Press, ??? (1979)

    1979

  10. [18]

    Journal of Physics A: Mathematical and General 29(7), 1511–1513 (1996) https://doi.org/10.1088/0305-4470/29/7/020

    Hu, G.Y., O’Connell, R.F.: Analytical inversion of symmetric tridiagonal matri- ces. Journal of Physics A: Mathematical and General 29(7), 1511–1513 (1996) https://doi.org/10.1088/0305-4470/29/7/020

    1996 doi

  11. [19]

    Journal of Combinatorial Theory, Series A 91(1-2), 191–214 (2000) https://doi.org/10.1006/jcta.2000.3094

    Chung, F., Yau, S.-T.: Discrete Green’s Functions. Journal of Combinatorial Theory, Series A 91(1-2), 191–214 (2000) https://doi.org/10.1006/jcta.2000.3094

    2000

  12. [20]

    Journal of Differential Geometry 23(2) (1986) https://doi.org/10

    Abresch, U., Langer, J.: The normalized curve shortening flow and homoth- etic solutions. Journal of Differential Geometry 23(2) (1986) https://doi.org/10. 4310/jdg/1214440025

    1986

  13. [21]

    Transac- tions of the American Mathematical Society, 5285–5309 (2012)

    Halldorsson, H.P.: Self-similar solutions to the curve shortening flow. Transac- tions of the American Mathematical Society, 5285–5309 (2012)

    2012

  14. [22]

    Nonlinearity 26(5), 1189–1226 (2013) https://doi

    Altschuler, D.J., Altschuler, S.J., Angenent, S.B., Wu, L.F.: The zoo of solitons for curve shortening in Rn. Nonlinearity 26(5), 1189–1226 (2013) https://doi. org/10.1088/0951-7715/26/5/1189

    2013 doi

  15. [23]

    Cambridge University Press, ??? (2012)

    Horn, R.A., Johnson, C.R.: Matrix Analysis, 2nd edn. Cambridge University Press, ??? (2012)

    2012

  16. [24]

    https://doi.org/10.1137/1

    Hartman, P.: Ordinary Differential Equations (2002). https://doi.org/10.1137/1. 9780898719222 . http://dx.doi.org/10.1137/1.9780898719222

    2002 doi

  17. [25]

    Journal of Differential Equations 247(9), 2620–2636 (2009) https://doi.org/10.1016/j.jde.2009.07.024

    Lin, Y.-C., Tsai, D.-H.: Evolving a convex closed curve to another one via a length-preserving linear flow. Journal of Differential Equations 247(9), 2620–2636 (2009) https://doi.org/10.1016/j.jde.2009.07.024

    2009 doi

  18. [26]

    In: Encyclopedia of Mathematics

    Viete theorem. In: Encyclopedia of Mathematics. European Mathematical Soci- ety (2023) 36

    2023

Pith tools

Reviewed August 27, 2026 · model on record in the stance chip above.