On two notions of curvature on singular surfaces

Maxime Marot

REVIEW 1 major objections 4 minor 1 cited by

On two notions of curvature on singular surfaces

T0 review · 1 major / 4 minor · reviewed 2026-08-03 · deepseek-v4-flash

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

Pith's one-line read Curvature-measure bounds on singular surfaces imply angle-comparison bounds, closing a missing direction in the theory.

desk verdict Likely true and important claim, but the proof has real gaps in the non-branching lemma and the strong-angle setup; deserves refereeing, not acceptance as is. read the letter →

arxiv 2511.08138 v4 pith:HQ2CYY3S submitted 2025-11-11 math.DG math.MG

Maxime Marot This is my paper · ORCID

classification math.DGmath.MG MSC 53C4551K1051F99

keywords boundedintegralcurvaturemeasureCBBCATRCD(κ2)singularsurfacesanglecomparisonsubharmonicmetrics

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 proves that on singular surfaces with bounded integral curvature, an inequality relating the curvature measure to the area measure is equivalent to the classical angle-comparison notion of curvature bound. The forward direction—showing that such a measure inequality implies the corresponding CBB or CAT condition—was a missing piece in the theory, known before only in the flat case. The proof works for arbitrary curvature bounds by exploiting an estimate that controls triangle area distortion in terms of the curvature measure. A corollary is that these surfaces satisfy the modern RCD(κ,2) condition in dimension two, connecting two generations of curvature-bound theories.

What carries the argument

The load-bearing machinery consists of the curvature measure ω and the area measure μ on a BIC surface, together with two identities: the Gauss–Bonnet formula for simple closed curves and an area-distortion estimate (Proposition 3) that bounds the difference between a triangle's area and the area of its Euclidean model triangle by a curvature-measure term. The proof combines these to show that the inequality ω≥κμ forces every triangle to have non-negative relative excess, which is then converted into angle comparison using the lower strong angle inequality (Proposition 2). Non-branching is established to guarantee that every triangle is homeomorphic to a disk.

What would settle it

Construct a BIC surface with ω≥κμ and no cusps that has a branching geodesic at a point with 0<ω({p})<2π; since the theorem implies such a surface must be CBB(κ) and non-branching, exhibiting one (or proving the needed regularity) would directly test the result.

Watch

Extended reading notes

Core claim

The central claim is made in Theorems A and B. Theorem A states that a BIC surface—a metric surface with locally bounded integral curvature—whose curvature measure ω satisfies ω≥κμ and which has no cusp (no point with curvature atom at least 2π) is CBB(κ). Theorem B states that if ω≤κμ with κ≤0, the surface is locally CAT(κ). The paper fills a gap in the theory by proving that these measure inequalities imply the corresponding angle-comparison bounds; the reverse implications were already available. The key step is to convert the measure inequality into a lower bound on triangle excess using the Gauss–Bonnet formula and an area-distortion estimate, then upgrade this to angle comparison via l

Load-bearing premise

The proof's derivation of angle comparison assumes that every sub-triangle of a BIC surface has well-defined strong angles and that branching points carry zero curvature atom; these regularities are asserted rather than proved.

Editorial extensions

If this is right

If the main theorem holds, every BIC surface with curvature measure bounded below by κ and no cusps is CBB(κ), and therefore satisfies RCD(κ,2) with its Hausdorff measure, by a known theorem.
The class of BICB(κ) surfaces without cusps then coincides with the class of complete CBB(κ) surfaces of Hausdorff dimension two.
For κ≤0, a curvature-measure upper bound implies local CAT(κ), and the local conclusion is optimal: a flat cylinder is CBB(0) and CAT(0) locally but not globally.
In the non-negative case, the no-cusp condition is automatically satisfied, so the theorem applies to all BICB(κ) surfaces with κ≥0.

Reading between the lines

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

The equivalence suggests that checking the measure inequality could become a practical test for synthetic curvature bounds on singular surfaces, potentially simpler than verifying angle comparisons directly.
The area-distortion technique is particular to two dimensions; whether an analogous estimate can upgrade measure bounds to synthetic bounds in higher dimensions is an open direction suggested by the proof.
The paper explicitly leaves open whether negative curvature bounds exclude cusps; a positive answer would remove the no-cusp hypothesis from the main theorem, while a counterexample would show the hypothesis is essential.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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 short version: this paper is an honest attempt to prove the long-sought implication BICB(κ) ⇒ CBB(κ) for all κ, with a new area-distortion argument that genuinely goes beyond Reshetnyak's κ=0 sketch. The algebra in Theorem 1 is coherent, and the paper is straight about what is new. But the proof as written has holes that need closing, so I would not accept it in current form.

The genuinely new piece is the estimate relating the area of a triangle to the area of its plane development, which lets them handle κ ≠ 0. The exposition of the background is clear, and the attribution to Reshetnyak and Petrunin is honest. There is no circularity or sign of fitted parameters; the logic of Theorem 1 itself hangs together.

The soft spots:

1. Lemma 2, the non-branching step, is load-bearing and its proof is not justified. The claim that at the first splitting point z the curvature measure vanishes, ω({z})=0, does not follow from 'no cusp'. Cusps only forbid ω({z}) ≥ 2π; smaller positive atoms are allowed. The turns of the two branches being negative does not force ω({z}) to be zero. So the conclusion that the sector angle is π is unsupported. Since Theorem A uses Lemma 2 to turn arbitrary triangles into disc-like ones, the main implication is incomplete as written.

2. Theorem 1 invokes Proposition 2, which requires strong angles to exist between the relevant sides, and does not verify or cite a reason that BIC surfaces have them. This may be fixable using deep results in Reshetnyak's theory, but it is not shown.

3. In the adjacent-hinge part of Theorem A, the paper says 'as p is an interior point of a geodesic, ω({p})=0'. I don't see why that is true; a cone point can lie in the interior of a minimizing geodesic. This needs an argument or a citation.

4. Theorem B is mostly delegated: the 'deformation' argument is by reference and the main comparison theorem is stated without proof. That is thin for a main theorem.

I think the central idea is sound and the paper is worth a serious referee. The right recommendation is: send it to review, but ask for a complete proof of non-branching, a justification of the strong-angle hypothesis, and a fuller treatment of the upper-curvature half. If those can be supplied, the paper would be a solid addition to the theory.

Referee Report

1 major / 4 minor

Summary. The paper claims to close a long-standing gap in the theory of surfaces of bounded integral curvature by proving that the curvature-measure inequalities ω ≥ κ μ and ω ≤ κ μ are equivalent to the Alexandrov curvature bounds CBB(κ) and CAT(κ), respectively. Theorem A states that a BIC surface with ω ≥ κ μ and no cusp is CBB(κ); Corollary 1 then derives RCD(κ,2) via Petrunin's theorem. Theorem B states that a BIC surface with ω ≤ κ μ and κ ≤ 0 is locally CAT(κ). The proofs are based on a local hinge-comparison theorem (Theorem 1 and 2), a non-branching lemma (Lemma 2), and uniqueness of geodesics (Lemma 3), together with an application of globalization theorems.

Significance. If the main theorems are correct, the paper fills a well-known gap in the comparison between two classical notions of curvature on singular surfaces and has an important consequence: BICB(κ) surfaces without cusps are RCD(κ,2) spaces. This would unify the Alexandrov and Reshetnyak perspectives and make the curvature-measure approach available for analytic applications. The claimed result is natural and the direction from Alexandrov to BIC was already known; the reverse direction is the missing piece. However, the current manuscript leaves several load-bearing steps insufficiently justified, so the significance is currently prospective rather than established.

major comments (1)

[§3.2, Lemma 3] In Lemma 3, the inequality 'the left-hand side is non-positive as the right-hand side is positive' is not justified. The left-hand side involves the turns κl(γ1)+κl(γ2)+ω(D); even under ω ≤ κ μ ≤ 0, the signs of the individual turns have not been established in the manuscript. The proof of uniqueness of geodesics is therefore not convincing as written. Since uniqueness is used in the proof of Theorem B, this is another load-bearing point.

minor comments (4)

[Abstract / Introduction] Typo: 'whithout' should be 'without'.
[§2.1, angle definitions] The notation [p x/y] for a hinge is used later in the paper but is not explicitly defined; the text defines a hinge as [p x/y] in the display but the prose sometimes writes [p x/y] and [p y/z] without clarifying the order of arguments. This can be confusing in the proof of Theorem A.
[References] Reference [4] is listed as 'A. D. Alexandrow, Über eine Verallgemeinerung der Riemannschen Geometrie 1 (1957), 33–84' — the journal or source is missing. Reference [7] is given as a preprint with an arXiv ID but no year in the text; please add full bibliographic data.
[§3.2, Theorem B proof] The phrase 'By the preceding lemma only branching phenomena can appear' should likely be 'no branching phenomena can appear', given Lemma 3's uniqueness statement. Please correct to avoid ambiguity.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the derivation reduces known BIC curvature-measure bounds to Alexandrov bounds via standard external theorems, not by assuming the conclusion or renaming fitted inputs.

full rationale

The central implication (Theorem A) starts from BICB(κ), defined by ω≥κμ, and aims to prove CBB(κ). The proof uses standard tools external to the paper: Gauss-Bonnet for BIC surfaces ([8, Ch. 13]), area distortion ([2, Theorem 3 p.262]), the isothermal representation ([8, Ch. 7]), Alexandrov's comparison propositions ([4]), and the globalization theorem ([3, Thm 8.31]). Theorem 1 derives an angle comparison from the curvature-measure inequality plus an area estimate; it does not assume the angle inequality it proves. Lemma 2's non-branching argument is asserted (and may contain a mathematical gap in concluding ω({z})=0), but that is a proof gap, not a circular reduction: it does not define BICB(κ) in terms of CBB, nor fit parameters to the target result. No fitted input is renamed as a prediction, and no load-bearing step is justified by a self-citation of the present author. The RCD corollary relies on Petrunin's external result. Thus the claimed derivation chain is not circular, even if its correctness is open.

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

The central claim rests entirely on the standard BIC apparatus (curvature measure, Gauss–Bonnet, area distortion, isothermal coordinates) and on Petrunin's RCD theorem; no free parameters or invented entities are introduced. The main burden is on the cited structural facts and on the unverified strong-angle hypothesis.

assumptions (6)

domain assumption Existence and local finiteness of the curvature measure ω on BIC surfaces
Definition 1 and [11, Theorem 2.18]; the paper builds all claims on this.
domain assumption Gauss-Bonnet formula ω(D)+κ_l(∂D)=2π for disc domains
Used in proofs of Theorem 1, Lemma 2, Lemma 3; cited to [8, Chapter 13].
domain assumption Area-distortion estimate (Proposition 3)
Central estimate in Theorem 1; cited to [2, Theorem 3 p.262].
standard math Lower strong-angle comparison (Proposition 2)
Quoted from [4]; assumed without verifying its hypothesis that strong angles exist.
standard math Globalization local CBB→CBB ([3, Theorem 8.31])
Used to conclude Theorem A.
domain assumption Petrunin's CBB(κ) ⇒ RCD(κ,2) for two-dimensional measure spaces
Used for the RCD corollary; external result [9].

how reviews work

0 comments

Cite this review

Pith. "Pith review of On two notions of curvature on singular surfaces." pith.science (2026). https://pith.science/paper/HQ2CYY3S

@misc{pith2026251108138,
  author       = {Pith},
  title        = {Pith review of: On two notions of curvature on singular surfaces},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HQ2CYY3S}},
  note         = {Machine review of arXiv:2511.08138}
}

abstract

In this paper, we investigate the equivalence of two distinct notions of curvature bounds on singular surfaces. The first notion involves inequalities of the form $\omega\geq\kappa\mu$ (resp. $\omega\leq\kappa\mu$) where $\omega$ is the curvature measure and $\mu$ the Hausdorff measure. The second notion is the classical Alexandrov curvature bound CBB (resp. CAT). We demonstrate that these two definitions are, in fact, equivalent. Specifically, we fill an important gap in the theory by showing that the inequalities imply the corresponding Alexandrov CBB (resp. CAT) bound. One striking application of our result is that, in combination with a result of Petrunin, the lower bound $\omega\geq\kappa\mu$ implies $RCD(\kappa, 2)$.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

Analysis on surfaces with locally bounded integral curvature
math.DG 2026-08 conditional novelty 8.0 of 10

Every complete BIC surface without cusps is infinitesimally Hilbertian, locally doubling, has a local Poincaré inequality, and admits a Hölder continuous heat kernel; a Dynkin condition on negative curvature upgrades ...

Pith tools

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