pith. sign in

arxiv: 2504.06956 · v2 · submitted 2025-04-09 · 🧮 math.PR · math-ph· math.MP

How does the supercritical GMC converge?

Pith reviewed 2026-05-22 20:35 UTC · model grok-4.3

classification 🧮 math.PR math-phmath.MP
keywords log-correlated Gaussian fieldsstar-scale invariant fieldsextremal pointsGaussian multiplicative chaosfreezing phenomenonsupercritical regimelocal structureconvergence
0
0 comments X

The pith

The law of the shape of star-scale invariant log-correlated Gaussian fields near extremal points is characterized, refining the freezing phenomenon in supercritical Gaussian multiplicative chaos.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper extends methods for log-correlated Gaussian fields to characterize the probabilistic law governing the configuration shape near extremal points of star-scale invariant fields. This characterization directly yields a more precise description of how the associated multiplicative chaos measure behaves in the supercritical regime. A sympathetic reader cares because the result explains the local geometry at the points that dominate the measure after the freezing transition occurs.

Core claim

We study the local structure of ⋆-scale invariant fields around their extremal points by characterising the law of the 'shape' of the field's configuration near such points. As a consequence, we obtain a refined understanding of the freezing phenomenon in supercritical Gaussian multiplicative chaos.

What carries the argument

The law of the shape of the field's configuration near extremal points, obtained for ⋆-scale invariant log-correlated Gaussian fields by extending Biskup & Louidor methods.

If this is right

  • The local configuration near extremal points follows a specific probability law derived from the scale-invariant structure.
  • Freezing in supercritical GMC is understood through the statistics of these local shapes rather than global properties alone.
  • Convergence questions for the supercritical measure can be reduced to properties of the extremal-point shapes.
  • The same shape law applies uniformly across the class of star-scale invariant fields studied.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The shape law may allow explicit computation of moments or tail probabilities for the GMC measure above the critical threshold.
  • Similar local-shape characterizations could apply to other log-correlated fields outside the star-scale invariant class if the Biskup-Louidor techniques extend.
  • The result suggests that convergence of the supercritical GMC occurs by the measure concentrating according to the identified shape distribution.

Load-bearing premise

The fields under study belong to the class of ⋆-scale invariant log-correlated Gaussian fields whose extremal local structure can be characterized following the methods of Biskup & Louidor.

What would settle it

A direct numerical sampling of field values in small neighborhoods of high points in a simulated star-scale invariant field that produces a distribution incompatible with the claimed shape law would falsify the characterization.

read the original abstract

In the spirit of [M. Biskup & O. Louidor, Adv. Math. 330 (2018)], we study the local structure of $\star$-scale invariant fields -- a class of log-correlated Gaussian fields -- around their extremal points by characterising the law of the "shape" of the field's configuration near such points. As a consequence, we obtain a refined understanding of the freezing phenomenon in supercritical Gaussian multiplicative chaos.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript extends the extremal-point analysis of Biskup & Louidor (2018) to the class of ⋆-scale invariant log-correlated Gaussian fields. It characterizes the law of the local 'shape' of the field configuration near extremal points and uses this characterization to obtain a refined description of the freezing phenomenon for supercritical Gaussian multiplicative chaos.

Significance. If the central claims are established, the work would strengthen the geometric understanding of supercritical GMC measures by providing an explicit description of the local field configuration around high points. This extends the subcritical results of Biskup & Louidor to a broader covariance class and supplies a concrete link between the shape law and the freezing transition, which is a load-bearing ingredient for convergence statements in the supercritical regime.

major comments (2)
  1. [§2 (main characterization) and §4 (freezing application)] The central derivation applies the Biskup & Louidor (2018) extremal characterization directly to ⋆-scale invariant fields. However, the covariance in the ⋆-scale invariant class satisfies scale invariance only up to a slowly varying function. It is not shown whether this perturbation preserves the tail of the maximum and the decorrelation structure between distant high points that are required for the explicit PPP description and the conditional field law. This assumption is load-bearing for the claimed shape characterization and for the subsequent freezing statement in the supercritical regime.
  2. [§3 (supercritical GMC section)] The normalization and conditioning used to extract the shape law are taken from the subcritical setting of Biskup & Louidor. In the supercritical regime the GMC measure is renormalized differently; the manuscript does not verify that the same conditioning on extremal points remains valid or that the resulting shape law is unaffected by the change in normalization.
minor comments (2)
  1. [§1.2] Notation for the slowly varying function in the covariance is introduced without an explicit reference to the precise regularity assumptions needed for the tail estimates.
  2. [Abstract] The abstract states the main results clearly but does not indicate the precise range of the slowly varying function for which the shape law holds.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and the detailed report. We address the two major comments point by point below. Clarifications will be added to the manuscript to make the required verifications explicit.

read point-by-point responses
  1. Referee: [§2 (main characterization) and §4 (freezing application)] The central derivation applies the Biskup & Louidor (2018) extremal characterization directly to ⋆-scale invariant fields. However, the covariance in the ⋆-scale invariant class satisfies scale invariance only up to a slowly varying function. It is not shown whether this perturbation preserves the tail of the maximum and the decorrelation structure between distant high points that are required for the explicit PPP description and the conditional field law. This assumption is load-bearing for the claimed shape characterization and for the subsequent freezing statement in the supercritical regime.

    Authors: We agree that an explicit check is needed. The ⋆-scale invariant class is defined so that the covariance equals the standard log-correlated form plus a slowly varying function L with L(t) = o(log t) as t → ∞. This perturbation affects only lower-order terms in the variance and does not change the leading logarithmic growth that determines the tail of the maximum. The decorrelation estimates between distant high points likewise carry over because the covariance difference remains bounded by a slowly varying term that vanishes in the relevant scaling limits. We will insert a short subsection (or remark) in §2 that records these estimates and confirms that the Biskup–Louidor arguments adapt verbatim once the o(log t) control is used. revision: yes

  2. Referee: [§3 (supercritical GMC section)] The normalization and conditioning used to extract the shape law are taken from the subcritical setting of Biskup & Louidor. In the supercritical regime the GMC measure is renormalized differently; the manuscript does not verify that the same conditioning on extremal points remains valid or that the resulting shape law is unaffected by the change in normalization.

    Authors: The shape law is a statement about the conditional law of the underlying Gaussian field given its values at the extremal points; it does not depend on the multiplicative renormalization that produces the GMC measure. The different normalization in the supercritical regime rescales the total mass of the measure but leaves the local field configuration around each peak unchanged. Consequently the same conditioning and the same shape law apply. We will add one paragraph in §3 that recalls this separation of the Gaussian field from the chaos measure and notes that the freezing description therefore inherits the shape law directly from §2. revision: yes

Circularity Check

0 steps flagged

No circularity; central characterization imported from independent external reference

full rationale

The paper explicitly follows methods from Biskup & Louidor (2018), an independent prior work by different authors published in Adv. Math. The abstract states the local structure is characterized 'in the spirit of' that reference for the ⋆-scale invariant class, then applies the result to supercritical GMC freezing. No self-citations, self-definitional reductions, fitted inputs renamed as predictions, or ansatzes smuggled via author-overlapping citations appear. The derivation chain relies on external benchmark rather than reducing to the paper's own inputs by construction. This is the normal case of an extension paper building on verified prior results.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No free parameters, axioms, or invented entities can be identified from the abstract alone.

pith-pipeline@v0.9.0 · 5585 in / 1020 out tokens · 37144 ms · 2026-05-22T20:35:39.451782+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.