An optimal $L^2$ extension for continuous $L^2$-optimal Hermitian metrics

Zhuo Liu

REVIEW 2 major objections 4 minor 17 references

An optimal $L^2$ extension for continuous $L^2$-optimal Hermitian metrics

T0 review · 2 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read This paper proves that continuous $L^2$-optimal Hermitian metrics on bounded planar domains satisfy an optimal $L^2$ extension theorem whose constant is $\pi$ divided by the squared logarithmic capacity, and consequently that all such…

desk verdict Genuinely new extension theorem with a clean Błocki-method proof; only real risk is the black-box equivalence imported from [7]. read the letter →

arxiv 2506.22840 v1 pith:JQFLYBMI submitted 2025-06-28 math.CV

Zhuo Liu This is my paper · ORCID

classification math.CV MSC 32U0532D1532L1532U35

keywords L2-optimalmetricsoptimalL2extensionGriffithssemi-positivitylogarithmiccapacitycomplexGreenfunctionsingularHermitianplanardomainsplurisubharmonicfunctions

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 establishes an optimal $L^2$ extension theorem for continuous $L^2$-optimal Hermitian metrics on bounded planar domains. It proves that whenever a Hermitian metric on a holomorphic vector bundle over such a domain is $L^2$-optimal, every fiber vector extends to a global holomorphic section whose $L^2$ norm is at most $\pi$ times the fiber norm divided by the squared logarithmic capacity of the domain at the point. From this extension statement the paper derives that every continuous $L^2$-optimal Hermitian metric is Griffiths semi-positive, settling open questions in [6] and [12]. The proof adapts the method of [3] for the sharp extension constant.

What carries the argument

The argument follows the method of [3] for proving optimal $L^2$ extension via the complex Green function. The central objects are the Green function $g_D(z, w)$ of the domain and the logarithmic capacity $c_D(w) = \exp(\lim_{z \to w}(g_D(z, w) - \log|z - w|))$. The proof constructs two convex functions $\tau$ and $\rho$ on the positive reals satisfying the identity $\bigl(1 - (\rho')^2/\tau''\bigr) e^{2\rho - \tau + t} = 1$, which is exactly what makes the final constant $\pi / c_D(w)^2$ appear. A modified weighted $L^2$ estimate (Proposition 2.2 and Theorem 2.3) supplies approximate solutions to the $\bar\partial$-equation, and a cut-off function built from those weights localizes the $\bar\partial$-closed form near the base point. A separate restriction lemma shows $L^2$-optimality is preserved when restricting to complex hyperplanes.

What would settle it

Find a bounded planar domain $D$, a continuous Hermitian metric $h$ on a trivial bundle over $D$, a point $w \in D$ and a fiber vector $s$ such that $(D, E, h)$ is $L^2$-optimal but every holomorphic section $f$ with $f(w) = s$ has $\int_D |f|^2_h \, d\lambda > \pi |s|^2_{h(w)} / c_D(w)^2$. Equivalently, exhibit a continuous $L^2$-optimal metric whose restriction to a complex line fails Griffiths semi-positivity, contradicting Theorem 1.5.

Watch

Extended reading notes

Core claim

The central claim, Theorem 1.4, is that if $(D, E, h)$ is $L^2$-optimal with $h$ continuous on a bounded planar domain $D$, then for every $w \in D$ and every $s \in E_w$ there exists $f \in H^0(D, E)$ with $f(w) = s$ and $\int_D |f|^2_h \, d\lambda \le \pi |s|^2_{h(w)} / c_D(w)^2$, where $c_D$ is the logarithmic capacity. This bound is sharp when $D$ is a disk. Theorem 1.5 then concludes that every continuous $L^2$-optimal Hermitian metric is Griffiths semi-positive, by restricting to complex lines and invoking the equivalence between the optimal $L^2$-extension property and Griffiths semi-positivity from [7, Theorem 1.3]. The proof also notes that only continuity at the base point is needed for the extension estimate, and that the triviality assumption on the bundle can be removed.

Load-bearing premise

The argument relies on the external equivalence, quoted from [7, Theorem 1.3], that for singular metrics on planar domains Griffiths semi-positivity is the same as the optimal $L^2$-extension property; if that equivalence requires regularity beyond upper semicontinuity, the passage from the extension theorem to Griffiths semi-positivity breaks.

Editorial extensions

If this is right

Continuous $L^2$-optimal metrics on planar domains satisfy an optimal extension inequality whose constant is the sharp one for disks.
Every continuous $L^2$-optimal Hermitian metric is Griffiths semi-positive, resolving Conjecture 1.3.
For direct image sheaves, if the input metric is $L^2$-optimal, the induced $L^2$-metric is continuous, locally $L^2$-optimal, and Griffiths semi-positive (Corollary 1.7).
The extension theorem holds under the weaker regularity condition of upper semicontinuity away from a closed pluripolar set.
The proof shows that the extension estimate only requires continuity of the metric at the base point $w$.

Reading between the lines

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

The same Green-function method may yield local optimal extension estimates for $L^2$-optimal metrics on higher-dimensional Stein manifolds by slicing along complex lines.
If the sharp constant is also sufficient, the optimal extension inequality for all small disks would characterize $L^2$-optimality, providing a converse to Theorem 1.4.
A direct test of the boundary of the method: weaken continuity to upper semicontinuity everywhere and check whether the extension constant still holds; Theorem 1.5's dependence on [7] suggests this may fail without extra assumptions.

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

The paper proves a genuinely new optimal L2 extension theorem: continuous L2-optimal Hermitian metrics on bounded planar domains admit holomorphic sections with the capacity constant bound. This was only known for smooth metrics or under an extension-property hypothesis. The proof follows Błocki's Suita-conjecture strategy and the displayed estimates are coherent; I did not find an internal error in Section 3. The applications are real: Theorem 1.5 answers Deng–Ning–Wang and Inayama, and Corollary 1.7 on direct images is a natural bonus. The author is also honest about the continuity hypothesis and gives a remark on when it can be relaxed.

The soft spots are mostly at the edges. Lemma 4.2 is quoted from [7, Theorem 1.3] and is load-bearing: it converts the disk-average extension property into Griffiths semi-positivity for singular metrics on planar domains. The paper states the lemma with upper semicontinuity assumptions that the continuous metrics here obviously satisfy, so the chain likely works. But a referee should check the original theorem statement in [7] to make sure it does not secretly require more regularity or a stronger extension property. If [7] is exactly as quoted, Theorem 1.5 follows; if not, the final implication breaks. The stress-test note flags this correctly as the central external dependency. I would also ask the author to state explicitly that L2-optimality passes to subdomains, which is silently used in Corollary 4.4; it is routine but not completely immediate.

There are a few display typos—(3.1) appears to drop a denominator and the section has some garbled Greek eta/theta—but they do not obscure the argument. The self-citation [13] for the restriction property is supported by a proof in Section 4, so that is fine.

This is a paper for SCV readers working on Ohsawa–Takegoshi, Suita, and positivity of singular metrics. It deserves a serious referee. I would send it to review rather than desk-reject. The proof is long but checkable, the new theorem is well-motivated, and the black-box dependency is isolated and verifiable.

Referee Report

2 major / 4 minor

Summary. The paper proves an optimal L^2 extension theorem (Theorem 1.4) for continuous L^2-optimal Hermitian metrics on bounded planar domains, with the sharp constant involving the logarithmic capacity. The proof follows Błocki's method and is carried out in detail. As an application, the author derives that every continuous L^2-optimal Hermitian metric on a holomorphic vector bundle is Griffiths semi-positive (Theorem 1.5), thereby resolving a conjecture of Deng–Ning–Wang and a question of Inayama, contingent on an external equivalence theorem from [7]. The paper also proves a restriction property for L^2-optimal metrics to complex hyperplanes (Lemma 4.3).

Significance. If the results are correct, the paper provides a sharp extension theorem under the strong global hypothesis of L^2-optimality and confirms a conjecture that was open for singular continuous metrics. The main analytic contribution is Theorem 1.4, which adapts Błocki's classical method with careful estimates and limiting arguments; the proof is detailed and the constant is explicit and parameter-free. The paper is not fully self-contained because the final implication relies on the external equivalence [7, Theorem 1.3], but the central new estimate appears sound. The restriction property (Lemma 4.3) is proved in the text and is a useful tool in its own right.

major comments (2)

[Section 4, Lemma 4.2 and Theorem 4.5] The proof of the main application, Theorem 1.5, depends entirely on Lemma 4.2, which is quoted from [7, Theorem 1.3] without proof. Since this lemma is a reformulation of a nontrivial equivalence (optimal L2-extension property on disks iff Griffiths semi-positivity) and is load-bearing for the resolution of Conjecture 1.3, the author should either provide a proof of the reformulation for planar domains or state precisely the original theorem and verify its hypotheses. In particular, confirm that upper semicontinuity of log|u|^2_{h*} is the only regularity needed, and that the disk-average condition is sufficient for the converse direction. If [7, Theorem 1.3] requires extra assumptions (e.g., local boundedness of h or a stronger extension property on all bounded pseudoconvex domains), the passage from Corollary 4.4 to Griffiths semi-positivity would be unjustified.
[Section 4, Corollary 4.4] The proof of Corollary 4.4 is incomplete as written. Theorem 1.4 applies only to bounded domains D that are themselves L2-optimal, but Corollary 4.4 asserts the optimal L2-extension condition for an arbitrary domain D based on applying Theorem 1.4 to disks D_r(w) ⋐ D. This requires the standard fact that L2-optimality restricts to relatively compact subdomains, which is not stated or proved in the manuscript. The author should add a lemma showing that if (D,E,h) is L2-optimal, then (Ω,E|_Ω,h|_Ω) is L2-optimal for every Ω ⋐ D; this is true by extending forms and weights by zero, but it needs to be explicit.

minor comments (4)

[Equation (3.1)] In the definition of α, the coefficient should be -χ' z/|z|^2 (or equivalently +χ' z/|z|^2 after accounting for the ordering of wedge products), not χ' \bar z/|z|^2. The subsequent estimates are unaffected because only the modulus-squared of the coefficient appears, but the formula should be corrected.
[Example 2.4] The formula for the complex Green function on the disk should have \bar w z in the denominator (i.e., g_{D_r}(z,w) = log | r(z-w)/(r^2 - \bar w z) |), and the logarithmic capacity formula should read c_{D_r}(w) = r/(r^2 - |w|^2), not r/(r^2 - |z|^2). The correct value c_{D_r}(0)=1/r is used later in Corollary 4.4.
[Section 2.1, proof of Theorem 2.3] In the displayed chain of inequalities after the definition of β, the term ⟨B^{-1}_φ β, β⟩ should read ⟨B^{-1}_λ β, β⟩, consistent with the application of Proposition 2.2.
[Section 4, Lemma 4.3] The proof of Lemma 4.3 is for restriction to a hyperplane. To obtain restriction to a complex line, as needed in the proof of Theorem 4.5, one must iterate the lemma. This is straightforward but should be stated explicitly.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the extension theorem is derived from the L2-optimality hypothesis via explicit ∂-estimates, and the Griffiths-positivity conclusion relies on an external theorem rather than on assuming the target claim.

full rationale

The paper's derivation chain is not circular. Theorem 1.4 is proved in Section 3 by taking the L2-optimality inequality (Definition 1.1) as the input, passing through Proposition 2.2 and Theorem 2.3 to solve the ∂-equation with weighted estimates, and then using the Blocki-type cutoff construction to extract a holomorphic section with the capacity-dependent bound. No step in this chain identifies the conclusion with the hypothesis by definition; the L2-optimality estimate and the final extension bound are genuinely different statements connected by a nontrivial proof. The final step, Theorem 1.5, uses Lemma 4.1, Lemma 4.2, Lemma 4.3, and Corollary 4.4. Lemma 4.2 is quoted from [7, Theorem 1.3], an external theorem that is not proved in the paper. Citing an external result is not circularity: it is independent support unless the external result reduces to the present target, and here [7] establishes an equivalence between Griffiths semi-positivity and an optimal extension property that the paper then verifies via Theorem 1.4 and Corollary 4.4. The only self-citation in the chain is [13, Proposition 3.3] for the restriction property, but the paper explicitly provides a full proof of this lemma in Lemma 4.3, so the self-citation is not load-bearing. Corollary 4.4 is a direct application of the disk capacity formula to Theorem 1.4 rather than a renaming of the hypothesis. Consequently, there is no step that reduces, by construction or by an unproved self-citation, to the paper's own claim. A possible concern about hidden regularity hypotheses in [7, Theorem 1.3] is a correctness risk, not a circularity finding, and it does not affect the circularity score.

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

The paper introduces no new objects or fitted constants; the auxiliary parameters epsilon, sigma, delta, c, a, b in the proof are regularization parameters that are sent to 0 or infinity and do not appear in the final statement. The central claim rests on standard potential theory and on one cited equivalence theorem ([7, Theorem 1.3]).

assumptions (5)

standard math Complex Green function exists and is unique for bounded planar domains, and the logarithmic capacity c_D(w) is positive.
Used throughout Section 3 to construct G_delta and the constant in Theorem 1.4.
standard math Richberg's global regularization theorem (Lemma 2.1, [17]) yields smooth strictly plurisubharmonic approximations of continuous plurisubharmonic functions.
Basis for Proposition 2.2, which extends L2-optimality from smooth weights to continuous weights with a lower bound lambda.
standard math Weak compactness, Mazur's theorem, and Fatou's lemma can be applied to extract limits of approximately holomorphic solutions.
Used in the limiting arguments in Sections 2 and 3 to pass from the regularized solutions u_epsilon to holomorphic sections.
domain assumption Equivalence of optimal L2-extension property and Griffiths semi-positivity for singular Hermitian metrics on planar domains (Lemma 4.2, cited from [7, Theorem 1.3]).
Load-bearing for Theorem 1.5: Theorem 1.4 provides the extension property, and this external result converts it to Griffiths semi-positivity.
standard math Griffiths semi-positivity can be checked by restriction to complex lines when log|u|^2_{h*} is upper semicontinuous (Lemma 4.1).
Used to reduce the higher-dimensional conjecture to the planar extension theorem.

how reviews work

0 comments

Cite this review

Pith. "Pith review of An optimal $L^2$ extension for continuous $L^2$-optimal Hermitian metrics." pith.science (2026). https://pith.science/paper/JQFLYBMI

@misc{pith2026250622840,
  author       = {Pith},
  title        = {Pith review of: An optimal $L^2$ extension for continuous $L^2$-optimal Hermitian metrics},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JQFLYBMI}},
  note         = {Machine review of arXiv:2506.22840}
}

abstract

In this paper, we obtain an optimal $L^2$ extension theorem for continuous $L^2$-optimal Hermitian metric on bounded planer domains. As applications, we affirmatively answer a question of Deng-Ning-Wang and a question of Inayama.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

17 extracted references · 17 canonical work pages

[7]

F. Deng, J. Ning, Z. Wang, X. Zhou: Positivity of holomorphic vector bundles in terms of Lp-estimates for ∂. Math. Ann. 385: 575–607 (2023)

work page 2023
[13]

Z. Liu, B. Xiao, H. Yang, X. Zhou: Multiplier submodule sheaves and a problem of Lempert. (2024) https://arxiv.org/abs/2111.13452v2

work page Pith review arXiv 2024
[1]

Berndtsson: A Brunn–Minkowski type inequality for Fano manifolds and some uniqueness theorems in K¨ ahler geometry, Invent

B. Berndtsson: A Brunn–Minkowski type inequality for Fano manifolds and some uniqueness theorems in K¨ ahler geometry, Invent. math.200 (2015), 149–200

work page 2015
[2]

Berndtsson, M

B. Berndtsson, M. P˘ aun: Bergman kernels and the pseudoeffectivity of relative canonical bundles. Duke Math. J. 145: 341–378 (2008)

work page 2008
[3]

B locki: Suita conjecture and the Ohsawa-Takegoshi extension theorem

Z. B locki: Suita conjecture and the Ohsawa-Takegoshi extension theorem. Invent. Math. 193(1): 149–158 (2013)

work page 2013
[4]

J. P. Demailly: Estimations L2 pour l’op´ erateur¯∂ d’un fibr´ e vectoriel holomorphe semi-positif au-dessus d’une vari´ et´ e k¨ ahl´ erienne compl` ete. Ann. Sci.´Ecole Norm. Sup. 15(3): 457–511 (1982)

work page 1982
[5]

J. P. Demailly: Complex analytic and differential geometry, (2012). Available at https: //www-fourier.ujf-grenoble.fr/ ~demailly/manuscripts/agbook.pdf

work page 2012
[6]

F. Deng, J. Ning, Z. Wang: Characterizations of plurisubharmonic functions. Sci. China. Math. 64(9): 1959–1970 (2021)

work page 2021

Show all 17 references

[8]

F. Deng, Z. Wang, L. Zhang, X. Zhou: New characterizations of plurisubharmonic functions and positivity of direct image sheaves. Amer. J. Math. 146(3): 751–768 (2024)

2024
[9]

Q. Guan, X. Zhou: A solution of an L2 extension problem with an optimal estimate and applications. Ann. of Math. 181(3): 1139–1208 (2015)

2015
[10]

Q. Guan, X. Zhou, L. Zhu: On the Ohsawa-Takegoshi L2 extension theorem and the twisted Bochner-Kodaira identity. C. R. Math. Acad. Sci. Paris 349: 797–800 (2011)

2011
[11]

H¨ ormander: L2 estimates and existence theorems for the ∂ operator

L. H¨ ormander: L2 estimates and existence theorems for the ∂ operator. Acta Math. 113: 89–152 (1965). 14 ZHUO LIU

1965
[12]

Inayama: Nakano positivity of singular Hermitian metrics and vanishing theorems of Demailly-Nadel-Nakano type

T. Inayama: Nakano positivity of singular Hermitian metrics and vanishing theorems of Demailly-Nadel-Nakano type. Algebr. Geom. 9(1): 69–92 (2022)

2022
[14]

Ohsawa, K

T. Ohsawa, K. Takegoshi: On the extension of L2 holomorphic functions. Math. Z. 195(2): 197–204 (1987)

1987
[15]

P˘ aun, S

M. P˘ aun, S. Takayama: Positivity of twisted relative pluricanonical bundles and their direct images. J. Algebraic Geom. 27: 211–272 (2018)

2018
[16]

Raufi: Singular hermitian metrics on holomorphic vector bundles

H. Raufi: Singular hermitian metrics on holomorphic vector bundles. Ark. Mat. 53(2): 359– 382 (2015)

2015
[17]

Richberg: Stetige streng pseudokonvexe funktionen

R. Richberg: Stetige streng pseudokonvexe funktionen. Math. Ann. 175(4): 257–286 (1967). Mathematical Science Research Center, Chongqing University of Technology, No. 69, Hongguang A venue, Banan District, Chongqing 400054, China. Email address : liuzhuo@cqut.edu.cn; liuzhuo@a...

1967

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [7]

F. Deng, J. Ning, Z. Wang, X. Zhou: Positivity of holomorphic vector bundles in terms of Lp-estimates for ∂. Math. Ann. 385: 575–607 (2023)

work page 2023

[2] [13]

Z. Liu, B. Xiao, H. Yang, X. Zhou: Multiplier submodule sheaves and a problem of Lempert. (2024) https://arxiv.org/abs/2111.13452v2

work page Pith review arXiv 2024

[3] [1]

Berndtsson: A Brunn–Minkowski type inequality for Fano manifolds and some uniqueness theorems in K¨ ahler geometry, Invent

B. Berndtsson: A Brunn–Minkowski type inequality for Fano manifolds and some uniqueness theorems in K¨ ahler geometry, Invent. math.200 (2015), 149–200

work page 2015

[4] [2]

Berndtsson, M

B. Berndtsson, M. P˘ aun: Bergman kernels and the pseudoeffectivity of relative canonical bundles. Duke Math. J. 145: 341–378 (2008)

work page 2008

[5] [3]

B locki: Suita conjecture and the Ohsawa-Takegoshi extension theorem

Z. B locki: Suita conjecture and the Ohsawa-Takegoshi extension theorem. Invent. Math. 193(1): 149–158 (2013)

work page 2013

[6] [4]

J. P. Demailly: Estimations L2 pour l’op´ erateur¯∂ d’un fibr´ e vectoriel holomorphe semi-positif au-dessus d’une vari´ et´ e k¨ ahl´ erienne compl` ete. Ann. Sci.´Ecole Norm. Sup. 15(3): 457–511 (1982)

work page 1982

[7] [5]

J. P. Demailly: Complex analytic and differential geometry, (2012). Available at https: //www-fourier.ujf-grenoble.fr/ ~demailly/manuscripts/agbook.pdf

work page 2012

[8] [6]

F. Deng, J. Ning, Z. Wang: Characterizations of plurisubharmonic functions. Sci. China. Math. 64(9): 1959–1970 (2021)

work page 2021