Analytic finite-rank corrections for singularly weighted estimates in a computer-assisted proof of 3D Euler singularity

Jiajie Chen; Thomas Y. Hou

REVIEW 2 major objections 5 minor 32 references

Analytic finite-rank corrections for singularly weighted estimates in a computer-assisted proof of 3D Euler singularity

T0 review · 2 major / 5 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read The paper establishes that numerical blowup profiles can be made to satisfy exact local vanishing conditions by low-rank analytic corrections, allowing singularly weighted energy estimates in a computer-assisted proof of 3D Euler singularit

desk verdict A clear, honest methods review of the authors' own finite-rank correction trick: the core mathematics is sound, the exposition is useful, and the main weakness is a conditional premise that rests on the published rigorous numerics of [13] plus a fixable typo in Eq. (4.24). read the letter →

arxiv 2607.15256 v1 pith:HJ5TTAT5 submitted 2026-07-16 math.AP cs.NAmath.NA

Jiajie Chen , Thomas Y. Hou This is my paper · ORCID

classification math.APcs.NAmath.NA MSC 35Q3135B4465G2076B03

keywords 3DEulersingularitycomputer-assistedproofsingularlyweightedestimatesfinite-rankcorrectionsvanishingorderBoussinesqequationsself-similarblowuprigorousnumerics

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 argues that the gap between numerical approximation and rigorous singular stability analysis in computer-assisted proofs of fluid blowup can be closed by a two-stage procedure: numerics fixes only coefficients of globally defined smooth basis functions, and then local constraints near the singularity are imposed analytically. The central requirement is that perturbations around the blowup profile vanish cubically, so that singular weights of order |x|^{-3} give finite energy norms. Raw numerical constructions preserve only quadratic vanishing, so the paper's method subtracts bad quadratic modes using Taylor expansions and explicit correction modes, making the improved vanishing order an exact property of the corrected function. The method is demonstrated on approximate space-time solutions and on the numerical stream function, and is presented as a reusable ingredient for other computer-assisted proofs.

What carries the argument

The central mechanism is the analytic low-rank correction: from a function f represented explicitly as a sum of basis functions, the mixed derivative ∂xyf(0) is defined exactly by applying derivatives to the basis; one then subtracts a correction mode with prescribed leading Taylor behavior (e.g., χ1=x+O(|x|^3), χ2=xy+O(|x|^3) with P(χ2)=0, χ3= -xy^2/2 κ*κ) to eliminate the bad quadratic modes. For space-time solutions, the correction amplitude a(t) solves the linear ODE (4.24)-(4.25) analytically, so the corrected function bg(2) has exact cubic vanishing of the residual. The decomposition W=W1+W2 and the residual operator R(W1) reformulate the equation exactly, so that only certified numeri

What would settle it

Take any raw numerical approximation, perturb its coefficients by a small noise, and apply the correction formulas (4.17), (4.21), (4.25). If the corrected function's mixed derivative at the origin (e.g., ∂xy(bg(2)-χ)|_{x=0}) is not identically zero—rather than merely small—for all perturbations, then the exactness claim is false. Concretely, evaluate the corrected residual's quadratic Taylor coefficient under exact arithmetic on the basis functions: a nonzero value would disprove the claim of analytic cubic vanishing.

Watch

Extended reading notes

Core claim

The central claim is that the improved vanishing order is an exact analytic property of the corrected function, even though the coefficients of the approximate functions are obtained numerically. Because the numerical step determines coefficients in explicit C^{4,1} basis representations (e.g., B-spline expansions), derivatives at the origin such as ∂xyF(0) are exact analytic quantities computed from the basis functions. Low-rank corrections subtract the leading Taylor defect modes, e.g., χ with χ=xy+O(|x|^3), and the correction coefficient a(t) is obtained by solving a linear ODE analytically. After correction, the residual error and the perturbation satisfy O(|x|^3) vanishing exactly, so t

Load-bearing premise

The assumption the whole argument leans on is that the numerical construction truly produces globally defined functions in an explicit smooth basis—not just grid values—so the Taylor coefficients used in the corrections are exact analytic quantities; if only approximate samples are available, the analytic cancellation collapses.

Editorial extensions

If this is right

The perturbation W1 in the weighted estimates has exact cubic vanishing, so the singularly weighted L∞ and C^{1/2} norms are finite and the claimed exponential decay holds.
The corrected approximate space-time solution bg satisfies bg(0,x)-χ=O(|x|^3) and (∂s-L)bg=O(|x|^3), making the residual operator a controllable small perturbation in the bootstrap.
The corrected stream function residual ε1 has cubic vanishing, so the velocity error u(ε1) satisfies the weighted functional inequality (5.3) and belongs to the energy space.
The methodology extends to higher vanishing order O(|x|^k) by adding more Taylor modes, and also to nonlocal PDEs with similar singularity structure.

Reading between the lines

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

Editorial: The same recipe—use numerics only to fix coefficients in a global basis, then perform all singular local analysis analytically—could be applied to enforce other exact local constraints such as boundary conditions, gauge choices, or compatibility conditions in computer-assisted proofs.
Editorial: Because the correction amplitude a(t) is small (of the order of the numerical error), the method tolerates approximate coefficient values; this suggests that interval arithmetic or certified rounding is needed only for the coefficient bounds, not for the vanishing-order cancellation itself.
Editorial: A transferable design rule emerges: any finite-rank operator chosen to approximate singular nonlocal terms can be paired with analytically constructed correction modes that have zero output under the nonlocal functional, decoupling the correction from the nonlocal estimate.
Editorial: One could turn the model problem into a standalone test: apply the correction to a known exact solution with deliberately noisy coefficients and check that the quadratic Taylor coefficient of the corrected function is identically zero, independent of the noise.

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

First thing to know: this is not a new-results paper. It is the authors explaining, in a clean model problem, the analytic finite-rank correction trick that made their computer-assisted singularity proof work. That is genuinely useful. The main claim — once numerical objects are coefficients in a fixed global C^{4,1} basis, Taylor corrections can enforce exact cubic vanishing near the origin — is correct, and the exposition in Section 4 is the clearest version I have seen. The stream-function correction in Section 5 is also coherent, and the explicit velocity formula for the corrected mode is a nice touch. Credit where due: the paper makes the bridge from numerical coefficients to analytic vanishing conditions transparent, and it does not oversell. The italic note correcting the typo in the early version of [11, Sec. 5.8] is exactly the kind of thing a review should do.

Soft spots, in proportion. First, the paper's central premise is conditional. Everything rests on the assertion that the actual numerical objects in [11,13] are exact global basis representations, with the origin away from patch boundaries and Taylor derivatives exactly computable from coefficients. The paper states this in Section 3, Step 1, but it does not prove it or supply code/data; it defers to [13]. Since [13] is published rigorous numerics, that is an acceptable deferral for a review, but the paper should say flatly that the premise is certified in [13], not just asserted. It is also silent on whether the certified coefficients are fixed reals or interval enclosures; if intervals, the word "exact" needs more nuance. That is a referee-level detail, not a fatal flaw. Second, Eq. (4.24) is malformed: a dangling "0=" and an inconsistent first equality. The intended ODE is clear from (4.25), so it is a typo, but in a paper whose purpose is clarity, it matters. Third, there is no new theorem here; the value is expository and the literature review is deliberately narrow.

For whom: people working on computer-assisted stability proofs, and anyone trying to understand or verify the Chen–Hou singularity proof. The reader's structural take is right. I would send this to peer review — not as a novel result but as a methods review — and ask for the typo fix plus an explicit statement about where and how the basis-representation premise is certified in [13]. With those changes, it deserves publication.

Referee Report

2 major / 5 minor

Summary. The paper is an expository/methods review that isolates a technique from the authors' earlier computer-assisted proof of singularity formation in the 2D Boussinesq and 3D axisymmetric Euler equations [11,13]. The technique consists of representing all numerical objects (approximate profiles, approximate space-time solutions, numerical stream functions) as exact finite linear combinations of smooth global basis functions, and then performing finite-rank analytic corrections based on Taylor expansions near the origin. These corrections enforce exact cubic vanishing conditions, making the corrected functions admissible for the singularly weighted energy estimates used in the stability proof. The paper reviews the stability framework, develops a scalar model problem in Section 4, and explains the corrections for the approximate space-time solution and the stream-function error. It does not prove a new theorem; its central claim is that the low-rank analytic correction principle is sound and is an essential component of the proof in [11,13].

Significance. If the method works as described, the paper is a valuable expository contribution: it isolates a non-obvious step in a major computer-assisted proof and presents it in a self-contained model problem. The distinction between numerical coefficients and globally-defined analytic functions is clearly made, and the Taylor-expansion cancellations in the model problem are mathematically coherent. The paper also gives a helpful overview of the overall proof structure and the role of singular weights and finite-rank perturbations. However, the paper proves no new theorem and provides no new numerical data or code. Its applicability to the actual proof rests on an explicit but unverified hypothesis about the basis representation of the numerical objects in [11,13], which is asserted rather than established here.

major comments (2)

[Sections 3 (Step 1), 4.2 (Step 1), 5 (Step 1)] The entire correction mechanism depends on the hypothesis that the numerical objects (e.g., \hat g^{(0)} in (4.15), \bar\phi^{N,(0)} in Section 5) are exact finite linear combinations of global C^{4,1} basis functions, with the origin in a region where a single representation (4.3a) applies and derivatives at the origin are exactly expressible from the coefficients. The paper asserts this repeatedly ('Once these coefficients are fixed, they determine functions defined on the entire domain') and defers to [13], but does not prove it or even state it as a precise assumption. If the mesh near the origin is adaptive, or if the coefficients are interval enclosures of a linear-system solution rather than fixed real numbers, then \partial_x\hat g^{(0)}(t,0) is not an exact quantity and the claimed cubic cancellation in (4.17), (4.21), and (5.1a) fails. Please state the precise hypothesis and gi
[Eq. (4.24)] Equation (4.24) is malformed. As printed, it reads '0 = A = B + \partial_{xy}E(s,0)', which would imply both A=0 and B+\partial_{xy}E(s,0)=0, and hence \partial_{xy}E(s,0)=0 unless the intended reading is nonstandard. The correct ODE, a'(s) - \lambda a(s) + \partial_{xy}E(s,0)=0, follows from (4.23) using \partial_{xy}(L\chi_2)(0)=\lambda, but this is not what is displayed. Since the explicit formula (4.25) for a(t) is a key output of the second correction step, this display must be corrected.

minor comments (5)

[Title/Abstract] The title contains an erroneous space: 'SINGULARL Y WEIGHTED'. Please fix.
[Section 5, Step 2] The function \chi_3 is defined as - (x y^2/2) \kappa_*(x)\kappa_*(y), but the text says '\kappa_*(y)=1+O(|y|^4)' while the x-variable condition on \kappa_* is not stated. Please state the condition for both variables.
[Eq. (5.3)] The notation '\rho_{10} \sim |x|^{-3}, \varphi \sim |x|^{-1/2}|x|^{-2}' is confusing; presumably the second expression means |x|^{-5/2}. Please clarify.
[Section 2.3, Lemma 2.1] In the statement 'md_2(\varphi_0) \leq -D 1_{R_1\leq |x|\leq R_2}', the indicator function notation should be defined or explained to avoid confusion with an ordinary function.
[Section 4.3, Example 1] The sentence 'By optimizing these two estimates, we obtain sharp piecewise bounds' is vague. For a self-contained exposition, at least a sketch of the optimization criterion would be helpful.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the corrected vanishing order is produced by explicit Taylor subtractions, not by fitting a prediction to its own target.

full rationale

The derivation under review is the analytic low-rank correction construction, not a new empirical prediction. In the model problem, the corrected function bg^(1) is defined in (4.17) by subtracting the exact first Taylor term ∂x bg^(0)(t,0)χ1, and bg^(2) is defined in (4.21) with a(t) chosen from the explicit analytic ODE (4.24)-(4.25) so that the mixed quadratic defect ∂xyE0(0) and ∂xyE(s,0) are cancelled. The vanishing order O(|x|^3) is therefore a construction criterion, not a fitted parameter renamed as a prediction. The paper repeatedly emphasizes that the correction uses exact analytic derivatives of basis-represented functions, e.g. (4.8), rather than numerical evaluations, so the cancellation is definitional in a legitimate mathematical sense, not circular. The substantial self-citations to [11,13] are contextual: Theorem 1.1 is explicitly stated as 'summarized from [11,13]', and the paper says it does 'not attempt to reproduce the entire stability proof'. This is an admitted scope limitation of a review, not a circular derivation. The remaining weakness—that the real 3D Euler numerical objects are not demonstrated here to be exact global basis representations—is a rigor/verification concern about antecedents, not circularity. The typographical dangling '0=' in (4.24) is a typo; the intended ODE follows from (4.22) and (4.23). No circular step can be exhibited.

Assumptions & free parameters 6 free parameters · 6 assumptions · 1 invented entities

The central method rests on numerically computed coefficients and on the assumption that numerical objects are actually represented by exact global smooth basis functions with certified bounds. The correction modes are explicit constructions, not new forces or entities. The main unshipped cost is the rigorous numerical machinery of [13], which is neither reproduced nor made publicly available here.

free parameters (6)

profile leading Taylor coefficients a1≈0.5, a2≈5.5, a3≈0.5 = a1≈0.5, a2≈5.5, a3≈0.5
Determined from the numerically constructed profile; used to set singular weight exponents and to motivate the model stability analysis (1.10)-(1.16).
B-spline coefficients c_i and a_{k,ij} = floating-point coefficients from numerical solver
Define the approximate profile, space-time solution bg, and stream function; all analytic corrections act on the resulting globally defined functions.
defect derivatives ∂xyE0(0) and ∂xyE(s,0) = small numerical-error-size values, not given explicitly
Enter the analytic ODE solution a(t) and enforce cubic vanishing in (4.19)-(4.25); their smallness is asserted from numerical error control.
stream-function residual derivatives ∂xε(0)(0) and ∂xyε(0) = not given explicitly
Determine the first and second low-rank corrections of the stream function in Section 5, equations (5.1)-(5.5).
singular-weight exponents α, β, m and the cubic weight |x|^{-3} = chosen by stability inequalities, e.g. γ > a3/a1, α > α*, m large
Hand-designed parameters of the weighted estimates; they are not predicted by the equations but selected to produce damping.
ODE growth rate λ̄ = c̄(0)−∂xV1(0)−∂yV2(0) = reported as < −5 for the actual Boussinesq system
Controls the exponential decay of a(t) in the correction; its value comes from the numerical profile and is used in (4.25).

assumptions (6)

domain assumption The 2D Boussinesq system (1.1) is the correct local approximation of the 3D axisymmetric Euler equations near the boundary singularity.
The entire stability analysis and Theorem 1.1 depend on this reduction, referenced to [26,10,11]; the paper states it without proof.
domain assumption Numerically determined coefficients define exact C^{4,1} functions on the whole domain via global basis representations.
Section 3 Step 1 and (4.15)-(4.16); analytic corrections require exact Taylor coefficients of these represented functions.
ad hoc to paper The rigorous numerical error estimates of [13, Appendices C,D] are correct and the computations are bug-free.
No code or certificates are provided in this paper; all smallness and bounds for defect coefficients are assumed from self-authored [13].
domain assumption The singularly weighted energy framework is valid: weight φ∼|x|^{-3} plus cubic vanishing gives bounded norms, and the outgoing condition (2.4) yields damping.
The correction method only matters because of this framework; Lemmas A.1/A.2 and (2.4) are cited from [11].
domain assumption The sharp Hölder estimates via optimal transport (Lemma 2.2) and the finite-rank decomposition hold with computable constants.
Section 2.4 relies on [11, Lemma 3.6] and [32]; these estimates are used to control nonlocal velocity terms.
standard math Local well-posedness and the Beale-Kato-Majda continuation criterion allow propagating the qualitative regularity needed in the bootstrap.
Section 4.5 uses the BKM-type criterion to close the logical cycle (a)-(e) for the full stability argument.

invented entities (1)

Analytic correction/cutoff modes χ, χ1, χ2, χ3, χε
purpose: Subtract bad quadratic Taylor modes and create explicit low-rank corrections so corrected functions vanish cubically near x=0.
These are explicitly constructed auxiliary functions (e.g. χ2=−Δ(1/6 xy^3 κκ), χ3=−xy^2/2 κ*κ*), not empirically testable entities; they are proof machinery, not physical or theoretical postulates.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Analytic finite-rank corrections for singularly weighted estimates in a computer-assisted proof of 3D Euler singularity." pith.science (2026). https://pith.science/paper/HJ5TTAT5

@misc{pith2026260715256,
  author       = {Pith},
  title        = {Pith review of: Analytic finite-rank corrections for singularly weighted estimates in a computer-assisted proof of 3D Euler singularity},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HJ5TTAT5}},
  note         = {Machine review of arXiv:2607.15256}
}

read the original abstract

Computer-assisted proofs of self-similar singularity formation for fluid equations often rely on numerically constructed approximate profiles. One effective approach to establishing stability of perturbations around a numerically constructed profile is to perform weighted energy estimates with singular weights near the singularity. However, the weighted norms require exact local vanishing conditions that are not automatically preserved by the equations nor the numerical construction. In this paper, we review an analytic low-rank correction method first developed in [ChenHou2023a,ChenHou2023b] to overcome this difficulty. The numerical step determines coefficients, rigorous bounds, and low-order defect modes in explicit global basis representations, while the required vanishing conditions are enforced analytically through low-rank corrections derived from Taylor expansions of the relevant quantities represented in a smooth basis. For completeness, we briefly review the singularly weighted estimates and a quantitative finite-rank perturbation method in the 2D Boussinesq / 3D Euler stability argument, where singular weights and the required vanishing order arise. Against this background, we formulate the local correction principle in a simplified setting, explain the correction of the residual error in numerical constructions of approximate space-time solutions and the stream function, and discuss its broader applicability to computer-assisted stability analysis for nonlocal PDEs.

Figures

Figures reproduced from arXiv: 2607.15256 by the authors.

**Figure 1.** Level 2: main steps. Level 3: main estimates and methods. Level 4: compute and estimate explicit functions or constants. 2¯cω + ¯ux ≈ −5.5, 2¯cω − u¯x ≈ 0.5, near x = 0. Thus ξ enjoys stronger damping than η. Moreover, since ∂xv¯(x, 0) = 0, ξ is weakly coupled with (ω, η) subsystem (2.2) near the boundary. 2.2. Steps of blowup analysis. The nonlinear stability estimates of the approximate blowup profile rely on thre… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 10 linked inside Pith

[13]

Stable nearly self-similar blowup of the 2D Boussinesq and 3D Euler equations with smooth data II: Rigorous numerics.Multiscale Modeling & Simulation, 23(1):25–130, 2025

Jiajie Chen and Thomas Y Hou. Stable nearly self-similar blowup of the 2D Boussinesq and 3D Euler equations with smooth data II: Rigorous numerics.Multiscale Modeling & Simulation, 23(1):25–130, 2025

2025
[1]

Finite time singularities in the landau equation with very hard potentials.arXiv preprint arXiv:2602.05981, 2026

Jacob Bedrossian, Jiajie Chen, Maria Pia Gualdani, Sehyun Ji, Vlad Vicol, and Jincheng Yang. Finite time singularities in the landau equation with very hard potentials.arXiv preprint arXiv:2602.05981, 2026

arXiv 2026
[2]

Smooth imploding solutions for 3d com- pressible fluids

Tristan Buckmaster, Gonzalo Cao-Labora, and Javier G´ omez-Serrano. Smooth imploding solutions for 3d com- pressible fluids. InForum of Mathematics, Pi, volume 13, page e6. Cambridge University Press, 2025

2025
[3]

Singularity formation and global well-posedness for the generalized Constantin–Lax–Majda equation with dissipation.Nonlinearity, 33(5):2502, 2020

Jiajie Chen. Singularity formation and global well-posedness for the generalized Constantin–Lax–Majda equation with dissipation.Nonlinearity, 33(5):2502, 2020

2020
[4]

On the slightly perturbed De Gregorio model onS 1.Arch

Jiajie Chen. On the slightly perturbed De Gregorio model onS 1.Arch. Ration. Mech. Anal., 241(3):1843–1869, 2021

2021
[5]

Remarks on the smoothness of theC 1,α asymptotically self-similar singularity in the 3D Euler and 2D Boussinesq equations.Nonlinearity, 37(6):065018, 2024

Jiajie Chen. Remarks on the smoothness of theC 1,α asymptotically self-similar singularity in the 3D Euler and 2D Boussinesq equations.Nonlinearity, 37(6):065018, 2024. 24 JIAJIE CHEN AND THOMAS Y. HOU

2024
[6]

Vorticity blowup in compressible Euler equations inR d, d≥3.Annals of PDE, 11(2):21, 2025

Jiajie Chen. Vorticity blowup in compressible Euler equations inR d, d≥3.Annals of PDE, 11(2):21, 2025

2025
[7]

Asymptotically self-similar blowup for 3D incompressible Euler withC 1,1/3− velocity I:C ∞ 1D limiting profiles.arXiv preprint arXiv:2605.15149, 2026

Jiajie Chen. Asymptotically self-similar blowup for 3D incompressible Euler withC 1,1/3− velocity I:C ∞ 1D limiting profiles.arXiv preprint arXiv:2605.15149, 2026

arXiv 2026

Show all 32 references

[8]

Asymptotically self-similar blowup for 3D incompressible Euler withC 1,1/3− velocity II: 3D profiles, blowup, and limiting behavior.arXiv preprint arXiv:2605.15130, 2026

Jiajie Chen. Asymptotically self-similar blowup for 3D incompressible Euler withC 1,1/3− velocity II: 3D profiles, blowup, and limiting behavior.arXiv preprint arXiv:2605.15130, 2026

2026 arXiv
[9]

Vorticity blowup in 2D compressible Euler equa- tions.To appear in Duke Math

Jiajie Chen, Giorgio Cialdea, Steve Shkoller, and Vlad Vicol. Vorticity blowup in 2D compressible Euler equa- tions.To appear in Duke Math. J., arXiv preprint arXiv:2407.06455, 2024

2024 arXiv
[10]

Finite time blowup of 2D Boussinesq and 3D Euler equations withC 1,α velocity and boundary.Communications in Mathematical Physics, 383(3):1559–1667, 2021

Jiajie Chen and Thomas Y Hou. Finite time blowup of 2D Boussinesq and 3D Euler equations withC 1,α velocity and boundary.Communications in Mathematical Physics, 383(3):1559–1667, 2021

2021
[11]

Stable nearly self-similar blowup of the 2D Boussinesq and 3D Euler equations with smooth data I: Analysis.arXiv preprint: arXiv:2210.07191v3 [math.AP], 2022

Jiajie Chen and Thomas Y Hou. Stable nearly self-similar blowup of the 2D Boussinesq and 3D Euler equations with smooth data I: Analysis.arXiv preprint: arXiv:2210.07191v3 [math.AP], 2022

2022 arXiv
[12]

Singularity formation in 3d euler equations with smooth initial data and boundary.Proceedings of the National Academy of Sciences, 122(27):e2500940122, 2025

Jiajie Chen and Thomas Y Hou. Singularity formation in 3d euler equations with smooth initial data and boundary.Proceedings of the National Academy of Sciences, 122(27):e2500940122, 2025

2025
[14]

On the finite time blowup of the De Gregorio model for the 3D Euler equations.Communications on Pure and Applied Mathematics, 74(6):1282–1350, 2021

Jiajie Chen, Thomas Y Hou, and De Huang. On the finite time blowup of the De Gregorio model for the 3D Euler equations.Communications on Pure and Applied Mathematics, 74(6):1282–1350, 2021

2021
[15]

Asymptotically self-similar blowup of the Hou–Luo model for the 3D Euler equations.Annals of PDE, 8(2):24, 2022

Jiajie Chen, Thomas Y Hou, and De Huang. Asymptotically self-similar blowup of the Hou–Luo model for the 3D Euler equations.Annals of PDE, 8(2):24, 2022

2022
[16]

On the stability of blowup solutions to the complex Ginzburg-Landau equation inR d.Annals of PDE, 11(2):29, 2025

Jiajie Chen, Thomas Y Hou, Van Tien Nguyen, and Yixuan Wang. On the stability of blowup solutions to the complex Ginzburg-Landau equation inR d.Annals of PDE, 11(2):29, 2025

2025
[17]

Smooth and stable euler implosions.arXiv preprint arXiv:2605.00808, 2026

Jiajie Chen, Steve Shkoller, and Vlad Vicol. Smooth and stable euler implosions.arXiv preprint arXiv:2605.00808, 2026

2026 arXiv
[18]

Blow-up for the incompressible 3D-Euler equations with uniform C 1, 1 2 −ϵ ∩L 2 force.arXiv preprint arXiv:2309.08495, 2023

Diego C´ ordoba and Luis Mart ´ ınez-Zoroa. Blow-up for the incompressible 3D-Euler equations with uniform C 1, 1 2 −ϵ ∩L 2 force.arXiv preprint arXiv:2309.08495, 2023

2023 arXiv
[19]

Finite time singularities to the 3D incompressible Euler equations for solutions inC ∞(R3 \ {0})∩C1,α ∩L 2.Annals of PDE, 11:19, 2025

Diego C´ ordoba, Luis Martinez-Zoroa, and Fan Zheng. Finite time singularities to the 3D incompressible Euler equations for solutions inC ∞(R3 \ {0})∩C1,α ∩L 2.Annals of PDE, 11:19, 2025

2025
[20]

Finite-time singularity formation forC 1,α solutions to the incompressible Euler equations on R3.Annals of Mathematics, 194(3):647–727, 2021

Tarek M Elgindi. Finite-time singularity formation forC 1,α solutions to the incompressible Euler equations on R3.Annals of Mathematics, 194(3):647–727, 2021

2021
[21]

On the stability of self-similar blow-up forC 1,α solutions to the incompressible Euler equations onR 3.arXiv preprint arXiv:1910.14071, 2019

Tarek M Elgindi, Tej-Eddine Ghoul, and Nader Masmoudi. On the stability of self-similar blow-up forC 1,α solutions to the incompressible Euler equations onR 3.arXiv preprint arXiv:1910.14071, 2019

1910 arXiv
[22]

From instability to singularity formation in incompressible fluids

Tarek M Elgindi and Federico Pasqualotto. From instability to singularity formation in incompressible fluids. arXiv preprint arXiv:2310.19780, 2023

2023 arXiv
[23]

Thomas Y Hou, Van Tien Nguyen, and Yixuan Wang.l 2-based stability of blowup with log correction for semilinear heat equation.arXiv preprint arXiv:2404.09410, 2024

2024
[24]

Hou, Yixuan Wang, and Changhe Yang

Thomas Y. Hou, Yixuan Wang, and Changhe Yang. Nonuniqueness of Leray-Hopf solutions to the unforced incompressible 3D Navier-Stokes equation.arXiv preprint arXiv:2509.25116, 2025

2025
[25]

Are the incompressible 3d navier–stokes equations locally ill-posed in the natural energy space?Journal of Functional Analysis, 268(12):3734–3766, 2015

Hao Jia and Vladimir Sverak. Are the incompressible 3d navier–stokes equations locally ill-posed in the natural energy space?Journal of Functional Analysis, 268(12):3734–3766, 2015

2015
[26]

Potentially singular solutions of the 3D axisymmetric Euler equations.Proceedings of the National Academy of Sciences, 111(36):12968–12973, 2014

Guo Luo and Thomas Y Hou. Potentially singular solutions of the 3D axisymmetric Euler equations.Proceedings of the National Academy of Sciences, 111(36):12968–12973, 2014

2014
[27]

Cambridge University Press, 2002

AJ Majda and AL Bertozzi.Vorticity and incompressible flow, volume 27. Cambridge University Press, 2002

2002
[28]

On the implosion of a compressible fluid ii: singularity formation.Ann

Frank Merle, Pierre Rapha¨ el, Igor Rodnianski, and Jeremie Szeftel. On the implosion of a compressible fluid ii: singularity formation.Ann. of Math. (2), 196(2):779–889, 2022

2022
[29]

Self-similar blow-up solutions of incompressible euler equations inR d, d≥3 withC 1,1−2/d− velocity.arXiv preprint arXiv:2605.19716, 2026

Feng Shao, Dongyi Wei, Ping Zhang, and Zhifei Zhang. Self-similar blow-up solutions of incompressible euler equations inR d, d≥3 withC 1,1−2/d− velocity.arXiv preprint arXiv:2605.19716, 2026

2026 arXiv
[30]

Global regularity of axisymmetric euler equations without swirl in higher dimensions.Acta Mathematica Sinica, English Series, 42(3):663–679, 2026

Feng Shao, Dongyi Wei, and Zhifei Zhang. Global regularity of axisymmetric euler equations without swirl in higher dimensions.Acta Mathematica Sinica, English Series, 42(3):663–679, 2026

2026
[31]

Incompressible Euler blowup at theC 1,1/3 threshold.arXiv preprint arXiv:2603.10945, 2026

Steve Shkoller. Incompressible Euler blowup at theC 1,1/3 threshold.arXiv preprint arXiv:2603.10945, 2026

2026 arXiv
[32]

American Mathematical Soc., 2021

C´ edric Villani.Topics in optimal transportation, volume 58. American Mathematical Soc., 2021

2021

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.

[1] [13]

Stable nearly self-similar blowup of the 2D Boussinesq and 3D Euler equations with smooth data II: Rigorous numerics.Multiscale Modeling & Simulation, 23(1):25–130, 2025

Jiajie Chen and Thomas Y Hou. Stable nearly self-similar blowup of the 2D Boussinesq and 3D Euler equations with smooth data II: Rigorous numerics.Multiscale Modeling & Simulation, 23(1):25–130, 2025

2025

[2] [1]

Finite time singularities in the landau equation with very hard potentials.arXiv preprint arXiv:2602.05981, 2026

Jacob Bedrossian, Jiajie Chen, Maria Pia Gualdani, Sehyun Ji, Vlad Vicol, and Jincheng Yang. Finite time singularities in the landau equation with very hard potentials.arXiv preprint arXiv:2602.05981, 2026

arXiv 2026

[3] [2]

Smooth imploding solutions for 3d com- pressible fluids

Tristan Buckmaster, Gonzalo Cao-Labora, and Javier G´ omez-Serrano. Smooth imploding solutions for 3d com- pressible fluids. InForum of Mathematics, Pi, volume 13, page e6. Cambridge University Press, 2025

2025

[4] [3]

Singularity formation and global well-posedness for the generalized Constantin–Lax–Majda equation with dissipation.Nonlinearity, 33(5):2502, 2020

Jiajie Chen. Singularity formation and global well-posedness for the generalized Constantin–Lax–Majda equation with dissipation.Nonlinearity, 33(5):2502, 2020

2020

[5] [4]

On the slightly perturbed De Gregorio model onS 1.Arch

Jiajie Chen. On the slightly perturbed De Gregorio model onS 1.Arch. Ration. Mech. Anal., 241(3):1843–1869, 2021

2021

[6] [5]

Remarks on the smoothness of theC 1,α asymptotically self-similar singularity in the 3D Euler and 2D Boussinesq equations.Nonlinearity, 37(6):065018, 2024

Jiajie Chen. Remarks on the smoothness of theC 1,α asymptotically self-similar singularity in the 3D Euler and 2D Boussinesq equations.Nonlinearity, 37(6):065018, 2024. 24 JIAJIE CHEN AND THOMAS Y. HOU

2024

[7] [6]

Vorticity blowup in compressible Euler equations inR d, d≥3.Annals of PDE, 11(2):21, 2025

Jiajie Chen. Vorticity blowup in compressible Euler equations inR d, d≥3.Annals of PDE, 11(2):21, 2025

2025

[8] [7]

Asymptotically self-similar blowup for 3D incompressible Euler withC 1,1/3− velocity I:C ∞ 1D limiting profiles.arXiv preprint arXiv:2605.15149, 2026

Jiajie Chen. Asymptotically self-similar blowup for 3D incompressible Euler withC 1,1/3− velocity I:C ∞ 1D limiting profiles.arXiv preprint arXiv:2605.15149, 2026

arXiv 2026