Extensions of One-Sided Box Geometry and Pyramid Invariants to gd-Sets and qm-Spaces

Shigeaki Yokota

REVIEW 2 major objections 3 minor 12 references

Extensions of One-Sided Box Geometry and Pyramid Invariants to gd-Sets and qm-Spaces

T0 review · 2 major / 3 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read This paper establishes that approximate domination is closed under box convergence, and that high-mass approximate directed maps force the target representation into every subsequential weak limit of the source pyramids.

desk verdict A serious directed analogue of Shioya's pyramid machinery with clean limit formulas; the load-bearing steps depend on same-author preprints, so the paper deserves refereeing but needs those imports made independently checkable. read the letter →

arxiv 2608.12749 v1 pith:PJQGDJ2N submitted 2026-08-13 math.MG math.PR

Shigeaki Yokota This is my paper · ORCID

classification math.MGmath.PR MSC 53C2354E3528A33

keywords geometricdatasetquasi-metricmeasurespaceboxdistancepyramidobservablediameterseparationsemi-LipschitzfunctionsLevyfamily

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 extends measured geometry to asymmetric and function-family settings, where a chosen family of functions, rather than the full Lipschitz class, defines the structure. Its central claim is that approximate domination is closed under box convergence: if each source dominates a target up to a vanishing additive error while both converge in box distance, then the limit source genuinely dominates the limit target. For directed spaces, approximate 1-Lipschitz maps defined only on Borel subsets of measure tending to one, with pushforwards converging weakly to the target measure, are shown to place the target representation in every subsequential weak limit of the source pyramids. For weakly convergent pyramids, observable diameter and multi-directed separation are recovered by perturbing their mass parameters and taking lower or upper limits, with both limits agreeing. This matters because finite measurement data from nearly all of the space, without exact global maps or symmetry, are enough to determine the limiting directed geometry.

What carries the argument

The load-bearing machinery is the one-sided box distance and the finite measurements attached to function families. The relation $Y\preceq_{\varepsilon} X$ means that a coupling $\pi$ and a closed relation $S$ carry mass at least $1-\varepsilon$, and every function of $Y$ is within $\varepsilon/2$ in sup norm on $S$ of some function of $X$; the one-sided box distance $\square_{\preceq}(Y,X)$ is the infimum of such $\varepsilon$. Limits are read through ordered finite measurements $M(X;N,R)$, the Prokhorov-closed sets of pushforwards of clipped $N$-tuples of functions, together with a reconstruction theorem: an $L$-gd-set belongs to a pyramid once its finite measurements are contained in the pyramid's. For qm-spaces the class $\mathrm{Lip}_1^+(X)=\{f:f(y)-f(x)\le d_X(x,y)\}$ represents the directed distance, and an inf-convolution extension lemma, a semi-Lipschitz version of McShane--Whitney, fills in maps defined only on high-mass subsets. The standing monoidal closure $TB\subset L$ identifies closed gd-sets with the compact pyramid class to which the paper's compactness and weak-convergence criteria apply.

What would settle it

A concrete way to settle the main claim is to find qm-spaces $X_n$ and a target $Y$ that satisfy every hypothesis of Theorem 3.11 but for which $\mathrm{Rep}_+(Y)$ is absent from some weak limit of the pyramids of $\mathrm{Rep}_+(X_n)$, or to compute an explicit weakly convergent pyramid sequence for which the right-perturbed upper limit in equation (22) differs from $\mathrm{ObsDiam}(P;-\kappa)$, which would disprove Theorem 4.5.

Watch

Extended reading notes

Core claim

At the center is a stability theorem for one-sided box geometry. The one-sided box distance $\square_{\preceq}(Y,X)$, the infimum of the additive error with which $Y$ is dominated by $X$, is shown to satisfy a triangle inequality and to vanish exactly when $Y\preceq X$; therefore domination is closed under box convergence. For qm-spaces, the paper proves that Borel maps $p_n$ from sets $A_n\subset X_n$ with $\mu_n(A_n)\to 1$ to $Y$, whose pushforwards $(p_n)_*(\mu_n|_{A_n})$ converge weakly to $\nu$ and whose directed distance distortion satisfies $d_Y(p_n(x),p_n(y))\le d_n(x,y)+\varepsilon_n$ with $\varepsilon_n\to0$, force $\mathrm{Rep}_+(Y)$ to belong to every subsequential weak limit of the pyramids generated by $\mathrm{Rep}_+(X_n)$. For weakly convergent pyramids, the exact limit formulas are established: $\mathrm{ObsDiam}(P;-\kappa)$ equals, as $\varepsilon\downarrow0$, both the lower and upper limits of $\mathrm{ObsDiam}(P_n;-(\kappa+\varepsilon))$, and $\mathrm{Sep}(P;\kappa_0,\ldots,\kappa_N)$ equals both limits of $\mathrm{Sep}(P_n;\kappa_0-\varepsilon,\ldots,\kappa_N-\varepsilon)$. In the general unclosed setting, observable diameter, nonnegative separation, and the two one-sided median-tail masses are compared, and a sequence is a function-family Levy family exactly when both median-tail masses vanish at every positive radius.

Load-bearing premise

The load-bearing premise is that the companion preprints' theorems are correct, namely that every closed gd-set is compact and that weak convergence of pyramids can be detected through finite measurements, and the proof transfers its results through those theorems without reproducing their proofs.

Editorial extensions

If this is right

If Theorem 3.11 is right, a target can be certified as belonging to a weak limit by constructing approximate 1-Lipschitz maps on sets of measure tending to one; exact global maps are unnecessary.
If Theorems 4.5 and 4.9 are right, the limiting observable diameter and separation distance of any weakly convergent pyramid sequence are computable from bounded finite-tuple measurement distributions, with the directional mass perturbations built into the formulas.
Domination being box-closed means that convergence in box distance cannot lose the relation 'which function family approximates which'; approximate relations pass to exact ones in the limit.
Without closure assumptions on the function family, the Section 5 inequalities hold for every gd-set, requiring neither constants, truncations, nor inf-convolutions among the observables.
The equivalent description of function-family Levy families says that uniform closeness to constants around the whole family is the same as both one-sided median-tail masses vanishing at every positive radius.

Reading between the lines

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

As an extension beyond the paper, the same perturbation mechanism should apply to any directed invariant that is monotone in its mass parameters, not only observable diameter and separation; the proofs use only monotonicity plus finite-measurement approximation.
A testable prediction beyond the paper is that the one-sided concentration functions satisfy limit formulas under weak pyramid convergence parallel to Theorem 4.5, because they are suprema over the same one-sided Lipschitz observables.
The high-mass subset theorem could become a numerical recipe: sample growing random subsets of a candidate target, simulate maps on them, check the pushforward measure and the directed distance distortion, and certify pyramid membership; this recipe is my inference, not a claim of the paper.
For symmetric mm-spaces the same formulas should reduce to the classical limit statements, so the asymmetric perturbations are best read as a directed refinement of the existing theory; this comparison is my inference.

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

Short version: this is a real extension, not a repackaging. Yokota introduces a one-sided box distance for gd-sets, proves it metrizes approximate directed domination, and shows that high-mass approximate 1-Lipschitz maps push the target into every weak subsequential limit of the source pyramids. Theorems 4.5 and 4.9, giving right-perturbation limits for observable diameter and left-perturbation limits for multi-directed separation, are the payoffs; the Section 5 characterization of function-family Levy families via two tail functions is a useful standalone result. Credit where due: the proofs carried out here are mostly detailed and checkable. Lemma 3.10, the one-sided McShane-Whitney extension with measure control, is nice, and the squeeze estimates in 4.5 and 4.9 are coherent. The paper's explicit list of every imported statement from [10]-[12] is good scholarly hygiene.

The soft spots are real but proportionate. First, Proposition 3.3: U is defined as the projection of a closed set, and projections of closed sets need not be closed. The line "Continuity gives the corresponding non-strict estimate on U" wants an argument or a slightly different definition. I think this is a small gap, not a fatal one. The larger issue is the one the stress-test flags: Lemmas 3.12-3.14 and Appendix A lean on [11, Thm 3.19, Prop 7.2, Lemmas 4.8/4.9] and [12, Thm A.7/Cor A.12], all same-author preprints whose proofs are not reproduced here. If the compactness theorem in [11] has hidden hypotheses that arbitrary qm-spaces do not satisfy, Theorems 3.11, 4.5 and 4.9 collapse. I did not find an independent reason to think the imports fail, and the author is unusually transparent about them, but transparent dependency is still dependency. A referee should be able to inspect [11] and [12] alongside, or the author should move the needed statements into this paper.

The citation pattern is fine: earlier symmetric work is credited, and self-citations point to the specific statements used.

Who is this for? People working in metric measure geometry, concentration of measure, and irreversible Finsler spaces. It deserves a serious referee. I would not desk-reject. I would send it out and ask for the Proposition 3.3 closure fix and either inclusion or independent verification of the companion theorems before acceptance.

Referee Report

2 major / 3 minor

Summary. This paper develops a one-sided version of Gromov--Shioya box geometry for geometric data sets (gd-sets) and quasi-metric measure spaces (qm-spaces). It defines a one-sided box distance \square^{\preceq}, proves its basic order properties and closure under box convergence (Theorem 1.1(i), Propositions 3.2--3.5), and establishes a high-mass approximate domination transfer: Borel maps defined on asymptotically full-measure subsets, with pushforwards converging weakly and directed-distance error tending to zero, force the target representation into every subsequential weak limit of the source pyramids (Theorem 1.1(ii), Proposition 3.11). For weakly convergent pyramids, the paper proves limit formulas for observable diameter after a rightward mass perturbation (Theorem 4.5) and for multi-directed separation after a common leftward mass perturbation (Theorem 4.9). A final section gives elementary comparisons between observable diameter, nonnegative directed separation, and one-sided concentration functions, together with a characterization of function-family L\'evy families (Corollary 5.6), illustrated by a discrete example. The paper is explicitly built on three companion papers by the same author, and it contains a self-declared ledger of every imported statement from those papers.

Significance. If the companion framework is correct, the results are a substantive extension of concentration and compactness theory to directed and asymmetric structures. The main formulas are parameter-free in the sense that no fitted constants appear: the perturbations in Theorems 4.5 and 4.9 are dictated by the limiting argument, not by normalization. The paper is also unusually transparent about its external dependencies, listing each imported result from [10], [11], and [12]. Several proofs are detailed and checkable, notably the zero-set argument in Proposition 3.4 and the squeeze estimates in Propositions 4.5 and 4.9. The discrete example in Section 5 cleanly illustrates the difference between function-family invariants and underlying-space invariants. The significance is conditional: the central transfer theorems 3.11, 4.5, and 4.9 rest on compactness, weak-convergence detection, and reconstruction results imported from unpublished same-author preprints, and those imports are not proved or machine-checked in this manuscript.

major comments (2)

[§3, Proposition 3.3] The central transfer theorems depend on unverified statements from the companion preprints [11] and [12]. In particular, the sentence before Lemma 3.12 asserts that by [11, Theorem 3.19] every L-closed gd-set is L-compact; Lemma 3.12 uses this to prove compactness of M(P;N,R), and Lemma 3.14 uses [11, Proposition 7.2] to convert Hausdorff convergence of finite measurements into weak convergence of pyramids and [11, Lemmas 4.8–4.9] for reconstruction. Proposition 3.11 then invokes Lemma 3.14 to place the target in the limit pyramid, and Propositions 4.5 and 4.9 use the same finite-measurement transfer. Appendix A additionally relies on [12, Theorem A.7, Proposition A.8, Corollary A.12]. None of these statements is proved here, and the manuscript does not state the precise hypotheses of [11, Theorem 3.19] nor verify, for instance, that the function families arising from Rep_+(X_n) satisfy any uniform boundedness condition that L-compactness might require. If the compactness or weak-convergence detection in [11] fails in this setting, then the step 'Theorems 3.12 and 3.14 gives λ in M(P;N,R)' in the proof of Proposition 3.11 is invalid, and Theorems 4.5 and 4.9 collapse as well. The explicit ledger of imported results is commendable, but it does not supply the missing verification. I recommend that the author either reproduce the needed statements and proofs in an appendix or ensure that the companion papers are accepted and publicly available in their final form, and that the manuscript explicitly verify that the hypotheses of [11, Theorem 3.19] hold for the function families used here.
[§3, Proposition 3.3] In the triangle inequality proof, the set U = pr13({(x,y,z) | (x,y)∈S, (y,z)∈T}) need not be closed, since it is a projection of a closed subset of a noncompact product. The one-sided box relation in Definition 3.1 requires a closed witness set. The proof should replace U by its closure \bar{U}; the mass estimate passes to \bar{U} because \theta(\bar{U}) ≥ \theta(U), and the approximate inequality for h and f passes to the closure by continuity of the functions involved. As written, the proof is incomplete at this point, though the repair is local.

minor comments (3)

[Throughout] Several cross-references use inconsistent numbering: the introduction refers to 'Theorem 3.11' for the high-mass domination result, which is stated as Proposition 3.11; Proposition 3.8 refers to 'Theorem 3.1' where Definition 3.1 is meant; Proposition 3.11 refers to 'Theorem 3.10' where Lemma 3.10 is meant; and the proof of Proposition 3.11 says 'Theorems 3.12 and 3.14 gives' where both are lemmas.
[§4, Proposition 4.5] In the proof of Proposition 4.5, when applying Equation (20), the passage 'Letting R→+∞' could be made more explicit for the case of infinite observable diameter; the argument is standard but the monotone-convergence step is not spelled out.
[§5, Corollary 5.6] The proof of the implication (ii)⇒(iii) handles κ>1/2 by saying that no admissible pairs exist; since Sep+ is then zero by Definition 5.1, this is correct, but it would be clearer to state explicitly that the vacuous case is covered by the definition.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the limit formulas are proved from measurement-set convergence and epsilon-squeeze arguments, not from fitted parameters or self-referential definitions; the companion-preprint imports are disclosed one-way dependencies.

full rationale

The paper's derivation chain is not circular under the stated review rules. The main invariants are defined directly from gd-set data (Equations (3) and (10); Definitions 4.2 and 4.6), and Theorems 4.5 and 4.9 are established by epsilon-squeeze arguments using the Hausdorff convergence of finite-measurement sets (Lemmas 3.12–3.14), rather than by fitting parameters or by defining the limit object to equal the sequence limit. The proofs of Lemmas 3.12–3.14 and of Theorem 3.11 do import compactness, weak-convergence detection, and reconstruction results from the author's companion preprints [11] and [12]; the manuscript explicitly discloses this in its list of fixed upstream statements and in the sentence before Lemma 3.12. The dependence is one-way and is stated as such: 'The dependence on the companion paper [12] is one-way: its abstract pyramid transport results are used here, while they do not use the quantitative estimates proved in this paper.' Under hard rule 4, cited theorems with their own stated assumptions are independent evidence and do not by themselves raise the circularity score. Concerns about the correctness of [11, Theorem 3.19] or [11, Proposition 7.2] would be correctness risks, not circularity. No step in the paper reduces, by construction or by renaming, a target result to an input; in particular, no fitted constant is relabeled as a prediction and no ansatz is smuggled through citation. The result is therefore a self-contained derivation modulo clearly acknowledged upstream results, and the appropriate circularity score is 0.

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

There are no fitted numerical constants: the epsilon perturbations are limits along sequences, and the mass parameters kappa are variables of the invariants. The load-bearing content is imported structure: the gd-set and pyramid frameworks, the monoidal closure condition, and the companion-paper transfer results. These are disclosed but not verified in this manuscript; they account for the medium correctness risk.

assumptions (5)

domain assumption gd-set framework: nonempty function family F_X generates a complete separable metric and carries a full-support Borel probability measure.
Adopted from [10, Definition 3.1] at the opening of the introduction; all gd-set results are stated in this class.
domain assumption Pyramid axioms and weak convergence are those of [11, Definition 5.1 and Proposition 7.2]: nonempty, downward closed, directed, box-closed families; weak convergence is detected by unordered finite measurements.
Used throughout Section 3 and in Theorems 4.5 and 4.9; not re-proved here.
domain assumption Standing monoidal family assumption TB subset L, and L-closed gd-sets are L-compact via [11, Theorem 3.19].
Invoked immediately before Lemma 3.12 to identify pyramids in L circle D with the compact class of [11].
domain assumption The transfer theorem for pyramidal adjunctions, [12, Theorem A.7], and the claim that representation-reconstruction adjunctions are pyramidal, [12, Corollary A.12].
These power the pullback limit formulas in Appendix A, Theorem A.3; the manuscript cites them without reproducing proofs.
standard math Gluing lemma for couplings and Strassen's theorem for Prokhorov distance.
Used in Proposition 3.3 for gluing and in Lemma 3.12 for Strassen's theorem; standard results from [9] and [6].

how reviews work

0 comments

Cite this review

Pith. "Pith review of Extensions of One-Sided Box Geometry and Pyramid Invariants to gd-Sets and qm-Spaces." pith.science (2026). https://pith.science/paper/PJQGDJ2N

@misc{pith2026260812749,
  author       = {Pith},
  title        = {Pith review of: Extensions of One-Sided Box Geometry and Pyramid Invariants to gd-Sets and qm-Spaces},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PJQGDJ2N}},
  note         = {Machine review of arXiv:2608.12749}
}

read the original abstract

Domination of gd-sets, the relation recording which function family approximates which, is closed under box convergence. More generally, approximate 1-Lipschitz maps on asymptotically full-measure subsets, whose restricted measures converge to the target measure after pushforward, place the target representation in every subsequential weak limit of the source pyramids. For weakly convergent pyramids, bounded joint distributions of finite ordered tuples of observables recover both observable diameter, after rightward perturbation of its mass parameter, and the largest common forward gap among several positive-mass sets, after common leftward perturbation of their mass parameters. The lower- and upper-limit formulas agree as the perturbations vanish. Without closure assumptions on a gd-set's function family, we compare observable diameter, the nonnegative part of forward separation, and upper and lower median-tail masses. All observables become uniformly close in measure to suitable constants exactly when both median-tail masses vanish at every positive radius.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

12 extracted references · 5 canonical work pages

[10]

S. Yokota. Geometry of geometric data set I. to appear in Tsukuba Journal of Mathematics, arXiv:2603.23316

arXiv
[12]

S. Yokota. Pyramidal compactification of asymmetric metric measure spaces via adjoint trans- port. preprint, arXiv:2608.01145, 2026

work page Pith review arXiv 2026
[11]

S. Yokota. Geometry of geometric data set II: Pyramid. preprint, arXiv:2603.23325

arXiv
[1]

Cheng and Y

X. Cheng and Y. Feng. Concentration and relevant properties of Finsler metric measure manifolds, 2025

work page 2025
[2]

Esaki, D

S. Esaki, D. Kazukawa, and A. Mitsuishi. Convergence of cones of metric measure spaces and its application to Cauchy distribution.Int. Math. Res. Not. IMRN, 2025(18):rnaf292, 2025. arXiv:2402.14331

arXiv 2025
[3]

Ohta.Comparison Finsler Geometry

S. Ohta.Comparison Finsler Geometry. Springer Monographs in Mathematics. Springer, Cham, 2021

work page 2021
[4]

Ozawa and T

R. Ozawa and T. Shioya. Limit formulas for metric measure invariants and phase transition property.Math. Z., 280(3-4):759–782, 2015

2015
[5]

Romaguera and M

S. Romaguera and M. Sanchis. Semi-Lipschitz functions and best approximation in quasi- metric spaces.J. Approx. Theory, 103(2):292–301, 2000

2000

Show all 12 references

[6]

Shioya.Metric measure geometry, volume 25 ofIRMA Lectures in Mathematics and Theoretical Physics

T. Shioya.Metric measure geometry, volume 25 ofIRMA Lectures in Mathematics and Theoretical Physics. EMS Publishing House, Z¨ urich, 2016. Gromov’s theory of convergence and concentration of metrics and measures. 26 SHIGEAKI YOKOTA

2016
[7]

T. Shioya. Metric measure geometry: an approach to high-dimensional and infinite- dimensional spaces.Sugaku Expositions, 35(2):221–241, 2022. Translation of S¯ ugaku71 (2019), no. 2, 159–177

2019
[8]

Stojmirovi´ c

A. Stojmirovi´ c. Quasi-metric spaces with measure.Topology Proceedings, 28(2):655–671, 2004

2004
[9]

Villani.Optimal transport, volume 338 ofGrundlehren der mathematischen Wis- senschaften

C. Villani.Optimal transport, volume 338 ofGrundlehren der mathematischen Wis- senschaften. Springer-Verlag, Berlin, 2009

2009

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.

[1] [10]

S. Yokota. Geometry of geometric data set I. to appear in Tsukuba Journal of Mathematics, arXiv:2603.23316

arXiv

[2] [12]

S. Yokota. Pyramidal compactification of asymmetric metric measure spaces via adjoint trans- port. preprint, arXiv:2608.01145, 2026

work page Pith review arXiv 2026

[3] [11]

S. Yokota. Geometry of geometric data set II: Pyramid. preprint, arXiv:2603.23325

arXiv

[4] [1]

Cheng and Y

X. Cheng and Y. Feng. Concentration and relevant properties of Finsler metric measure manifolds, 2025

work page 2025

[5] [2]

Esaki, D

S. Esaki, D. Kazukawa, and A. Mitsuishi. Convergence of cones of metric measure spaces and its application to Cauchy distribution.Int. Math. Res. Not. IMRN, 2025(18):rnaf292, 2025. arXiv:2402.14331

arXiv 2025

[6] [3]

Ohta.Comparison Finsler Geometry

S. Ohta.Comparison Finsler Geometry. Springer Monographs in Mathematics. Springer, Cham, 2021

work page 2021

[7] [4]

Ozawa and T

R. Ozawa and T. Shioya. Limit formulas for metric measure invariants and phase transition property.Math. Z., 280(3-4):759–782, 2015

2015

[8] [5]

Romaguera and M

S. Romaguera and M. Sanchis. Semi-Lipschitz functions and best approximation in quasi- metric spaces.J. Approx. Theory, 103(2):292–301, 2000

2000