Some questions about the regularity and the uniqueness of solutions of parabolic partial differential equations

Inmaculada Gayte Delgado; Irene Mar\'in Gayte

REVIEW 3 major objections 3 minor 11 references

Some questions about the regularity and the uniqueness of solutions of parabolic partial differential equations

T0 review · 3 major / 3 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read A shape identity for linear parabolic equations is claimed to prove uniqueness of 3D Navier-Stokes weak solutions.

desk verdict A paper whose central theorem fails on an elementary heat-equation example, with independent errors in the Navier-Stokes section; not ready for review. read the letter →

arxiv 2505.22110 v1 pith:L5HAY3T3 submitted 2025-05-28 math.AP

Inmaculada Gayte Delgado , Irene Marín Gayte This is my paper

classification math.AP MSC 35Q3049K2076D0593C95

keywords Initial-boundaryvalueproblemsforsecond-orderparabolicequationsNavier-StokesuniquenessPDEsSmoothnessandregularityofsolutionstoWeakheatequationmaximumprinciplecontrollability

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 sets out to prove that the solution of a linear parabolic equation with a positive bounded source and nonnegative initial data is, at every final time, proportional to the solution of the heat equation started from the same data. From that shape identity it derives a representation of signed solutions as a difference of two heat solutions, and hence an $L^\infty(0,T;L^4(\Omega))$ bound for linear parabolic equations and systems. Applying that regularity to the nonlinear convective term $B(y,y)$ in the Navier-Stokes equations, the paper concludes that three-dimensional weak solutions are unique for initial data in $H$ and forcing in $L^2(0,T;V')$. If the chain of proofs is correct, this closes an open uniqueness question in fluid dynamics.

What carries the argument

The load-bearing mechanism is a final-time maximum principle (Theorem 3). If $\beta\in C^1([0,T])$ peaks only at $T$, and $w$ satisfies $w_t-\Delta w\ge 0$ with $w(0)\ge 0$, then the solution of $z_t-\Delta z=-\beta' w$ with nonpositive initial data is nonpositive at $T$; the proof uses a backward-heat test function and the sign of $F(\beta(t))$ to show $z(T)\le 0$. This sign control is iterated to produce the monotone source sequence $v_k$, with inequalities (11) and (12) keeping $0\le v_k\le u$, preserving $v_k=\varepsilon u$ on the cylinder, and forcing the final-time shape equality. The weak-limit and convex-combination step passes the equality to the limit, and the countable covering shrinks the support to nothing.

What would settle it

Solve the linear problem (1) on a bounded interval or square with $y_0$ the first Dirichlet eigenfunction and $u$ a positive constant, and compare the normalized profiles $y(T,x)/\|y(T)\|_{L^2}$ and $\varphi(T,x)/\|\varphi(T)\|_{L^2}$ at several times $T$; Theorem 4 says they coincide pointwise, so any measurable difference at a time when the heat profile is not flat would falsify the central identity.

Watch

Extended reading notes

Core claim

At the center of the paper is the identity $y(T)/\|y(T)\|_{L^2} = \varphi(T)/\|\varphi(T)\|_{L^2}$ for $0<c\le u\le M$ and $y_0\ge 0$, where $y$ solves $y_t-\Delta y=u$ with zero Dirichlet data and $\varphi$ solves the homogeneous heat equation with the same initial data. The paper proves this by constructing, for each small cylinder $\omega\times I$ and each $\varepsilon\in(0,1)$, a controlled source $v^*_\varepsilon$ with $0\le v^*_\varepsilon\le u$, $v^*_\varepsilon=\varepsilon u$ on $\omega\times I$, and the same final-time shape as $y$; a weak-limit and diagonal argument then drives the control to zero, leaving the pure heat profile $\varphi$. The identity is extended to signed data and general $u\in L^2(0,T;H^{-1}(\Omega))$ as $y(t)=\lambda_1(t)\varphi_1(t)-\lambda_2(t)\varphi_2(t)$, yielding the claimed $L^\infty(0,T;L^4)$ regularity for linear systems and, by splitting a Navier-Stokes solution into heat and linear parts, the claimed uniqueness in three dimensions.

Load-bearing premise

The argument hinges on the existence, for every small cylinder and every $\varepsilon\in(0,1)$, of a control $v^*_\varepsilon$ that stays between $0$ and $u$, equals $\varepsilon u$ on that cylinder, and makes the controlled heat flow have the same final-time shape as the forced solution $y$; if such a control fails to exist even once, the chain from Theorem 4 to the Navier-Stokes uniqueness breaks.

Editorial extensions

If this is right

Every weak solution of the three-dimensional Navier-Stokes problem with $y_0\in H$ and $f\in L^2(0,T;V')$ would be unique, not merely the smoother ones.
A linear parabolic solution with $L^4(\Omega)$ initial data and $H^{-1}$-valued forcing would lie in $L^\infty(0,T;L^4(\Omega))$, componentwise for systems, regardless of the sign of the source.
For nonnegative data, the pointwise direction $y(t)/|y(t)|$ would be independent of the source $u$; only the scalar factor $|y(t)|/|\varphi(t)|$ changes.
Uniqueness for initial data in $H\cap L^4$ would transfer to all of $H$ through the spectral approximation argument in Theorem 12.

Reading between the lines

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

Beyond the paper, the shape identity would imply that, for nonnegative data, the whole family of forced linear parabolic flows follows one trajectory in projective space determined solely by the initial profile; this is checkable numerically on a simple domain before trusting the Navier-Stokes conclusion.
Beyond the paper, the $L^\infty(0,T;L^4)$ gain for linear systems suggests a template for other quadratic PDEs whose energy estimates use $L^4$ norms; whether the convective term of related fluid models can be treated in the same way is a natural test.
Beyond the paper, the proof singles out the existence of the bounded controls $v^*_\varepsilon$ as the gate for the whole chain; a computational search for such controls on a single cylinder would de-risk the argument independently of Navier-Stokes.

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.

Desk Editor's Note

read the letter

Quick take: Theorem 4 is false. The counterexample is too simple to wave away: on Ω=(0,π), take y0=sin x and u≡1. The heat solution is φ(T)=e^{-T} sin x. The forced solution is y(T)=e^{-T} sin x + (1-e^{-T}) x(π−x)/2. The steady term has a nonzero sin 3x coefficient, namely 4(1-e^{-T})/(27π), so y(T) is not a scalar multiple of φ(T). Hence y(T)/|y(T)| ≠ φ(T)/|φ(T)|. That kills Theorem 4 and everything that depends on it: the derived regularity results in Corollary 2 and Theorems 5–9, and the Navier-Stokes uniqueness claims in Theorems 11–13.

What is actually new: the idea of using a final-time maximum principle and a controllability construction to force a pointwise proportionality between a forced linear solution and the heat solution. I don't know another paper doing that. The exposition is clear and the authors know the standard literature (Temam, Lions, Prodi, Caffarelli–Kohn–Nirenberg, Lin). The target regularity, L∞(0,T;L4), is genuinely the classical sufficient condition for Navier-Stokes uniqueness, so the ambition is reasonable.

Where it breaks: the load-bearing Step 1 of Theorem 4 asserts the existence of v*_ε with exact final-time proportionality. The counterexample shows that no such control can exist in general. The rest of the proof—convex combinations, diagonal covering, passing to limits—cannot repair a false base step. There are independent problems in Section 4. In Theorem 11, subtracting the heat solution z from a Navier-Stokes solution y gives w_t−νΔw = f−B(y,y)−∇p, and the pressure gradient is silently dropped. Theorem 12 uses ‖y(t)‖_{L4} for a weak solution before any L4 estimate is available, which makes that argument circular. The citation pattern is ordinary; I don't object to relying on [2] for Theorem 3, but the new step is exactly where the argument fails.

Bottom line: I would desk-reject this. A serious referee would only need the one-line counterexample to reject. The paper is worth keeping as a cautionary example of how a control-theoretic approach to Navier-Stokes uniqueness can look credible and still be wrong. If the authors can prove a genuine controllability statement, the project might be worth revisiting, but this version is not.

Referee Report

3 major / 3 minor

Summary. The paper claims to establish a pointwise fixed-point identity y(t)/|y(t)| = φ(t)/|φ(t)| for solutions of linear parabolic problems with positive data and positive bounded source, where φ solves the homogeneous heat equation with the same initial data. From this identity it derives L∞(0,T;L4) regularity for linear parabolic equations and then uniqueness of three-dimensional Navier-Stokes weak solutions for initial data in H and forces in L2(0,T;V′). The proof of the central identity is a controllability construction in the spirit of the authors' earlier work [2], using a final-time maximum principle and a diagonal covering argument. The paper concludes that these results imply uniqueness of 3D Navier-Stokes weak solutions.

Significance. If the central identity were true, the paper would resolve a major open problem in the theory of the Navier-Stokes equations, namely uniqueness of weak solutions in three dimensions. The paper also attempts to give a new regularity theorem for linear parabolic equations with L2(0,T;H−1) data. However, the central identity is false: an elementary one-dimensional heat-equation example contradicts Theorem 4 directly. Because Theorems 8–13 are all built on this identity, the claimed regularity and uniqueness results collapse. The paper has no machine-checked proofs or reproducible code; its main positive feature is the explicit use of a controllability construction, but that construction fails at its load-bearing step.

major comments (3)

[Theorem 4 (Section 2)] The central statement is false. Take Ω=(0,π), y0=sin x, and u≡1. The heat solution is φ(t)=e^{-t} sin x. The forced solution is y(t)=e^{-t} sin x + (1−e^{-t}) x(π−x)/2. The second term has a nonzero Fourier sine coefficient at sin 3x, namely 4(1−e^{-T})/(27π), while φ(T) is supported on sin x only. Hence y(T) is not a scalar multiple of φ(T), contradicting y(T)/|y(T)|=φ(T)/|φ(T)|. Since Step 1 of the proof asserts the existence of controls v*_ε with exactly this proportionality, the asserted controllability premise is false; the error propagates to Corollary 2 and Theorems 8–9.
[Theorem 8 (Section 3)] The passage from L∞ right-hand sides to u∈L2(0,T;H^{-1}) is not valid. In the proof, u1=u+|u|+c and u2=|u|+c are used after taking limits, but |u| is not defined for a general distribution u∈H^{-1}, and the map u↦|u| is not continuous on H^{-1}. The convergence of the approximating states therefore does not yield the limiting identity y(t)=λ1(t)φ1(t)−λ2(t)φ2(t) with the stated λ_i defined through |y_i(t)|/|φ_i(t)|.
[Theorem 11 (Section 4)] The reduction of a Navier-Stokes solution to a linear parabolic problem omits the pressure. Setting z as the heat solution and w=y−z, the strong form of the equation for w is w_t−νΔw = f−y·∇y−∇p, not f−B(y,y). In the weak formulation in V the pressure term disappears, but then the operator is the Stokes operator, and the scalar componentwise heat-equation regularity of Theorem 9 does not apply without additional justification. Thus Theorem 11's conclusion y∈L∞(0,T;L4) is unsupported.

minor comments (3)

[Throughout] There are numerous typographical errors, including 'I Rsastisfying' in Theorem 1, 'pression' for pressure, and inconsistent notation for the space V (the set of smooth divergence-free functions and its H^1_0 closure are both denoted V).
[Theorem 4, Step 1] The proof uses inequalities with sup and inf of w1 without specifying whether these are essential suprema/infima and without explaining rigorously why inf_{ω×I} w1>0; a precise justification is needed.
[References] Reference [6] misspells 'Caffarelli-Kohn-Nirenberg' as 'Cafarelli-Kohn-Nirenberg', and several references are incomplete or inconsistently formatted.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the proof chain is structurally self-contained; the falsehood of Theorem 4 is a correctness defect, not a circular dependence.

full rationale

The claimed derivation is structurally self-contained. Theorem 4, the load-bearing result, is proved by an explicit monotone iteration (equations (8)-(21)) rather than by assuming the target proportionality; the auxiliary final-time maximum principle in Theorem 3 is attributed to the authors' prior work [2] but is also proved in the text, so the self-citation is not load-bearing. The passage from the constructed controls v*_epsilon to v*=0 via weak convergence and a diagonal covering is a deduction, not an input. The later fixed-point representation y(t)=lambda_1(t)phi_1(t)-lambda_2(t)phi_2(t) and the L^infty(0,T;L^4) conclusions follow algebraically once Theorem 4 is accepted. There is no place where the desired equality y(T)/|y(T)|=phi(T)/|phi(T)| is inserted as an assumption or where a fitted parameter is relabeled as a prediction. This is not an endorsement of correctness: for Omega=(0,pi), y0=sin x and u=1, the explicit solution y(t)=x(pi-x)/2+e^{-t}(sin x-x(pi-x)/2) has a nonzero sin 3x mode at time T, while phi(T)=e^{-T}sin x does not, so Theorem 4 is false. But a false proof step is a mathematical error, not a circular dependence; the circularity score is therefore 0.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

No numerical parameters are fitted. The proof depends on standard parabolic facts and on two ad hoc premises: the self-cited final-time maximum principle ([2]) and the asserted controllability sequence in Theorem 4, which fails.

assumptions (5)

standard math Maximum principle for linear parabolic equations with Dirichlet boundary conditions.
Used implicitly to compare y and Ψ_v with different forcings and to assert positivity in Theorem 4 and Corollary 1.
standard math Interpolation inequality ∥v∥_{L4} ≤ 2^{1/2}∥v∥_{L2}^{1/4}∥∇v∥_{L2}^{3/4} for N=3 (Temam, Lemma 3.5).
Invoked in Theorem 12 to bound the trilinear term in the uniqueness argument.
domain assumption Existence of at least one weak solution of 3D Navier-Stokes and the classic uniqueness criterion for regular enough solutions (Temam [10]).
Used in Theorem 11 to convert L∞(0,T;L4) regularity into uniqueness.
ad hoc to paper Final-time maximum principle (Theorem 3) inherited from [2].
This lemma is the engine of the control construction in Theorem 4; it is a self-cited result and its application in the iteration is not independently verified.
ad hoc to paper For every ε,ω,I there exists a control v*ε with the required shape equality (Step 1 of Theorem 4).
This is the load-bearing premise that permits the limit argument. The one-dimensional counterexample shows it is false.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Some questions about the regularity and the uniqueness of solutions of parabolic partial differential equations." pith.science (2026). https://pith.science/paper/L5HAY3T3

@misc{pith2026250522110,
  author       = {Pith},
  title        = {Pith review of: Some questions about the regularity and the uniqueness of solutions of parabolic partial differential equations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/L5HAY3T3}},
  note         = {Machine review of arXiv:2505.22110}
}

read the original abstract

This work obtains a fixed-point equation for the solution of linear parabolic partial differential problems based on solutions to heat problems. This is a pointwise equality, so we have required non-standard techniques that involve the study of the sign of certain solutions to linear parabolic problems. This fixed-point equation implies regularity properties of solutions to parabolic problems, not necessarily linear, and this allows us to prove the uniqueness of the solution in three dimensions for the Navier-Stokes problem.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

11 extracted references · 11 canonical work pages

[2]

& Mar ´ın-Gayte, I

Gayte, I. & Mar ´ın-Gayte, I. 2025 A new method for the exact controllability of linear parabolic equations. Mathematics 13 (3), 344. 17

work page 2025
[1]

& Nirenberg, L

Caffarelli, L., Kohn, R. & Nirenberg, L. 1982 Partial regularity of suitable weak solutions of the Navier-Stokes equations. Communications on Pure and Applied Mathematics 35, 771–831

work page 1982
[3]

J. L. Guermond, Faedo-Galerkin weak solutions of the Navier-Stokes equations with Dirichlet boundary conditions are suitable, J. Math. Pure Appl., Vol. 88, 2007, 87-106

work page 2007
[4]

Heywood, R

J.G. Heywood, R. Rannacher, and S. Turek, Artificial boundaries and flux and pressure conditions for the incompressible Navier?Stokes equations , Int. J. Numer. Meth. Fluids 22 (1996), 325–352

work page 1996
[5]

Ladyzhenskaya, O. A. & Seregin, G. A. 1999 On partial regularity of suitable weak solutions to the three-dimensional Navier-Stokes equations. J. Mathematical Fluid Mechanics. textbf2

work page 1999
[6]

Lin, F. H. 1998 A new proof of the Cafarelli-Kohn-Nirenberg theorem. Comm. Pure Appl. Math. , 51, 241–257

work page 1998
[7]

Lions, J. L. 1963 Quelques m´ ethodes de r´ esolutions des probl` emes aux limites non lin´ eaires. Paris. Dunod

work page 1963
[8]

1959 Un teorema di unicita per le equazioni di Navier-Stokes

Prodi, G. 1959 Un teorema di unicita per le equazioni di Navier-Stokes. Annali di Mat. , 48, 173-182

work page 1959

Show all 11 references

[9]

H. Sohr, W. Wahl, On the regularity of the pressure of weak solutions of Navier- Stokes equations, Arch. Math., Vol. 46, 1986, 428-439

1986
[10]

1977 Navier-Stokes equations

Temam, R. 1977 Navier-Stokes equations. Theory and numerical analysis. Studies in Mathematics and Applications, 2. Amsterdam. North-Holland Publishing Co

1977
[11]

You have made us for Yourself, O Lord, and our heart is restless till it rests in You

Da Veiga, H. B. 1985 On the suitable weak solution to the Navier-Stokes equations in the whole space. J. Math. pures Appl. , 64, 77–86. “You have made us for Yourself, O Lord, and our heart is restless till it rests in You”. Saint Augustine. 18

1985

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [2]

& Mar ´ın-Gayte, I

Gayte, I. & Mar ´ın-Gayte, I. 2025 A new method for the exact controllability of linear parabolic equations. Mathematics 13 (3), 344. 17

work page 2025

[2] [1]

& Nirenberg, L

Caffarelli, L., Kohn, R. & Nirenberg, L. 1982 Partial regularity of suitable weak solutions of the Navier-Stokes equations. Communications on Pure and Applied Mathematics 35, 771–831

work page 1982

[3] [3]

J. L. Guermond, Faedo-Galerkin weak solutions of the Navier-Stokes equations with Dirichlet boundary conditions are suitable, J. Math. Pure Appl., Vol. 88, 2007, 87-106

work page 2007

[4] [4]

Heywood, R

J.G. Heywood, R. Rannacher, and S. Turek, Artificial boundaries and flux and pressure conditions for the incompressible Navier?Stokes equations , Int. J. Numer. Meth. Fluids 22 (1996), 325–352

work page 1996

[5] [5]

Ladyzhenskaya, O. A. & Seregin, G. A. 1999 On partial regularity of suitable weak solutions to the three-dimensional Navier-Stokes equations. J. Mathematical Fluid Mechanics. textbf2

work page 1999

[6] [6]

Lin, F. H. 1998 A new proof of the Cafarelli-Kohn-Nirenberg theorem. Comm. Pure Appl. Math. , 51, 241–257

work page 1998

[7] [7]

Lions, J. L. 1963 Quelques m´ ethodes de r´ esolutions des probl` emes aux limites non lin´ eaires. Paris. Dunod

work page 1963

[8] [8]

1959 Un teorema di unicita per le equazioni di Navier-Stokes

Prodi, G. 1959 Un teorema di unicita per le equazioni di Navier-Stokes. Annali di Mat. , 48, 173-182

work page 1959