Central limit theorems for non-linear functionals of Gaussian fields via Wiener chaos decomposition

Fabio Coppini; Wioletta M. Ruszel

REVIEW 3 major objections 3 minor 37 references

Central limit theorems for non-linear functionals of Gaussian fields via Wiener chaos decomposition

T0 review · 3 major / 3 minor · reviewed 2026-08-03 · deepseek-v4-flash

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

Pith's one-line read A field-level central limit theorem turns nonlinear functionals of stationary Gaussian lattice fields into Gaussian white noise, with an application to powers of the discrete Gaussian free field.

desk verdict The field-level Breuer-Major packaging is tidy, but the advertised odd-power GFF limit is off by a factor of N in the normalization, so the central application fails. read the letter →

arxiv 2512.13148 v2 pith:HZG3FNJJ submitted 2025-12-15 math.PR math-phmath.MP

Fabio Coppini , Wioletta M. Ruszel This is my paper

classification math.PRmath-phmath.MP MSC 60G6060G1560F0560F1733C45

keywords WienerchaosexpansionHermiterankcentrallimittheoremGaussianwhitenoisediscretefreefieldcontinuousfourthmomentSobolevspaces

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 sets out to prove a field-level central limit theorem for nonlinear functionals of stationary Gaussian fields on the integer lattice, generalizing the classical scalar CLT for such functionals. Its main theorem states that if the function H has Hermite rank m and the field's covariance is absolutely summable at power m, then the normalized field N^{d/2} Phi_N converges in the Sobolev space H^{-alpha}(D) to Gaussian white noise on the unit cube D. The proof works directly on the field: each Hermite component of H(X_j) is written as a multiple stochastic integral, the fourth moment theorem forces each component to a Gaussian limit, and a separate tightness estimate in H^{-alpha}(D) upgrades convergence of observables to convergence of the field. The paper applies this machinery to powers of the discrete Gaussian free field: even powers have Hermite rank 2 and converge to white noise once the covariance is square-summable (d >= 5, or d >= 2 for gradients), while odd powers, whose linear Hermite term cannot be removed, are claimed to converge to a continuous Gaussian free field with explicit Green-function covariance. The odd-power result rests on a Riemann-sum approximation of the discrete Green's function.

What carries the argument

The mechanism is the Wiener chaos decomposition: H(X_j) is expanded into Hermite polynomials H_q(X_j), and each term is represented as a multiple stochastic integral I_q(u_j^{⊗q}) with respect to the isonormal process W over the Hilbert space with inner product rho(j-k). The proof then invokes the fourth moment theorem, which says that for a sequence of variables living in a single Wiener chaos, convergence to a Gaussian is equivalent to the vanishing of the contractions f_N ⊗_r f_N; the paper checks this for the normalized kernels s_{N,q}(f) by splitting the covariance sum into small and large scales and applying a standard norm inequality. Tightness in H^{-alpha}(D) is obtained by expandin

What would settle it

Compute the left side of equation (19) for f ≡ 1 and H(x) = x^3 on a d = 3 box: using G(i,j) ~ c|i-j|^{2-d}, the leading term c_1^2 N^{-d} sum_{i,j in B_N} G(i,j) diverges like N^2, so the variance of N^{-d/2} sum_{j in B_N} X_j^3 grows with N; observing this growth for increasing N refutes the claimed convergence to a continuous Gaussian free field.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that a white-noise limit can be obtained simultaneously for all chaos components of a nonlinear Gaussian lattice field, not just for scalar sums. Concretely, for a stationary unit-variance Gaussian field X on Z^d and a function H with Hermite expansion starting at order m, the condition sum_u |rho(u)|^m < infinity implies N^{d/2} Phi_N converges in law to W in H^{-alpha}(D) for every alpha > d/2, where W is Gaussian white noise on D = [-1/2,1/2]^d. The paper further claims that for the discrete Gaussian free field, H(x) = x^{2p} falls under this theorem in d >= 5, whereas H(x) = x^{2p+1} escapes it because the covariance is not summable, and

Load-bearing premise

The odd-power theorem depends on the assumption that the Riemann sum c_1^2 N^{-d} sum_{i,j in B_N} f(i/N)f(j/N) G(i,j) converges to c_1^2 int int f(x) G_cont(x,y) g(y) dx dy; for the discrete Gaussian free field Green's function in d >= 3 the sum grows like N^2, so this premise is not met.

Editorial extensions

If this is right

The field-level statement is stronger than a scalar CLT: for any finite collection of test functions, the vector (⟨Phi_N,f_i⟩) converges to a Gaussian vector with covariance diagonal in Hermite order, and disjoint-box indicators give independent standard normal limits.
Even powers of the discrete Gaussian free field, normalized by N^{d/2}, converge to white noise for d >= 5; the gradient version extends the result to every d >= 2.
Odd powers of the discrete Gaussian free field are claimed to have a continuum Gaussian free field limit whose covariance kernel is the continuous Green's function; the linear Hermite term dominates and the higher-order Hermite terms vanish in the limit.
The tightness bound in H^{-alpha}(D) shows the convergence holds as random distributions, so the limit statement applies to nonlinear functionals evaluated against Sobolev test functions.

Reading between the lines

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

A direct computation of the leading term in the paper's key variance identity for odd powers and f ≡ 1, using the discrete Green's function G(i,j) ~ |i-j|^{2-d}, gives growth of order N^2 rather than convergence; if this is correct, the odd-power claim needs a different normalization or a different limiting object.
The same contraction technique could yield quantitative rates of convergence for the even-power white-noise limit by tracking how fast the fourth-moment gap decays.
For d = 3 and d = 4, even powers fall outside the theorem because sum_u |rho(u)|^2 diverges; whether a different scaling produces a Gaussian or non-Gaussian limit is left open.

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

Colleague,

Read this one with a careful eye on the scaling. The paper's main claim, Theorem 3, says odd powers of the discrete GFF, normalized by N^{d/2}, converge to a continuous GFF with Green's covariance. That cannot be right. The linear term in the variance at Eq. (19) involves N^{-d} Σ f(i/N)f(j/N)G(i,j). For the GFF Green's function G(i,j) ~ |i-j|^{2-d}, that sum behaves like N^2, not O(1). So the normalization would need an extra factor N^{-1}; as stated, the variance diverges. The stress-test note is correct: the Riemann-sum identity in Section 1.3.2 is simply false for the discrete Green's function. This is not a minor gap; it is the load-bearing step that carries Theorem 3.

Credit where it's due. The first part of the paper, the field-level formulation of Breuer-Major with tightness in Sobolev spaces, is a clean and honest repackaging of known tools. The proof via the Nualart-Peccati fourth-moment theorem is standard but competently executed. The even-power application (white noise for even powers of the GFF for d≥5) is a correct and useful observation, though it follows quickly from known results like [CHRR23] and the original Breuer-Major theorem.

There are two other soft spots worth flagging. Theorem 1 defines C_m with |rho(u)|^q but the limiting variance in Proposition 1 uses rho(u)^q. If the covariance is signed, those don't match, and the theorem as stated is false; the paper should either assume rho≥0 or replace the variance with the absolute-value version. And Lemma 4 states lim λ_k/k^2 = 1, which is the one-dimensional Weyl law; in d dimensions the eigenvalues grow like k^{2/d}. That's a careless misstatement, though the subsequent bound may still hold for the intended α > d/2, so it might not break the tightness argument.

Overall, the paper does a decent job of reviewing and repackaging known material, but its advertised new application is incorrect, and the general theorem has an unaddressed sign issue. A serious referee would catch these, and the paper deserves that scrutiny, but as it stands it's not publishable without a substantial revision that either fixes the odd-power normalization or drops that claim entirely.

Recommendation: reject in current form, but encourage the authors to resubmit with the odd-power section corrected or removed, and with the C_m sign issue resolved.

Would I bring it to reading group? Maybe, as a cautionary example of how a scaling error can look like a Riemann sum. Would I cite it? Not now.

Best, [Your name]

Referee Report

3 major / 3 minor

Summary. The paper develops a field-level Breuer–Major CLT for non-linear functionals of stationary Gaussian lattice fields. The proof represents H(X_j) via its Hermite expansion, proves a CLT for each Wiener chaos using the Nualart–Peccati fourth-moment theorem, and adds a tightness argument in Sobolev spaces H^{-α}(D). The main applications are to powers of the discrete Gaussian free field: even powers are claimed to converge to Gaussian white noise, while odd powers are claimed to converge to a continuous Gaussian free field with explicit covariance (17). The odd-power claim is the advertised central novelty. As written, the proof of that claim rests on a false Riemann-sum identity; in addition, the normalization in Theorem 1 is inconsistent with the variance computed in Proposition 1, and Lemma 4 contains a false eigenvalue asymptotic.

Significance. If the field-level formulation were correct, it would be a natural extension of the classical Breuer–Major theorem and would provide a unified proof of some known GFF fluctuation results, particularly for even powers. The exposition of the standard chaos machinery is clear, and the proof of Proposition 1 is a mostly standard contraction argument. However, the odd-power GFF theorem is the paper's principal new application and it is false as stated because of the scaling error in Eq. (19). The normalization mismatch in Theorem 1 also affects the correctness of the main theorem for signed covariances. The contribution as it stands is not established; substantial correction of the statements and proofs would be required.

major comments (3)

[§1.3.2, Eq. (19) / Theorem 3] The variance computation in Eq. (19) is not a Riemann-sum approximation. For the discrete GFF Green function on Z^d, G(i,j) ≍ |i-j|^{2-d}. Taking f≡1 gives N^{-d} ∑_{i,j∈B_N} G(i,j) ≍ N^{-d} · N^{d+2} = N^2, and for general f the expression grows like N^2 ∫∫ f(x)f(y)|x-y|^{2-d} dxdy. Hence the first term in (19) diverges, and no limit with covariance (17) is obtained at normalization N^{-d/2}; the required normalization for the linear term is N^{-d/2-1}. Since this identity is the structural premise of Theorem 3, the theorem as stated is false. Separately, the coefficient in Eq. (18), c_1 = E[X_o^{2p+2}], equals G(o,o)^{p+1}(2p+1)!!, not G(o,o)^p(2p+1)!!.
[Theorem 1 / Proposition 1] The normalization constant in Theorem 1 is inconsistent with the variance computation in Proposition 1. Theorem 1 defines C_m = ∑_{q≥m} q! c_q^2 (∑_u |ρ(u)|^q), but Proposition 1, Eq. (33), gives the limit variance q! c_q^2 (∑_u ρ(u)^q) ∫ f^2. Unless ρ≥0 pointwise, the normalized field has variance (∑ ρ^q)/(∑ |ρ|^q) ∫ f^2, not the claimed white-noise covariance ∫ f^2. The theorem needs an explicit nonnegativity assumption or a C_m built from ∑ ρ^q (with a separate positivity/convergence condition). As stated, the normalization cannot produce the claimed limit for signed covariances.
[Lemma 4 (tightness)] The statement lim_{k→∞} λ_k/k^2 = 1 is false for the Dirichlet Laplacian on D ⊂ R^d with d≥2; Weyl's law gives λ_k ≍ k^{2/d}. The sup-bound (54) and the subsequent summability in (59) remain true for α>d/2 once the correct eigenvalue growth is used, so the tightness argument is repairable, but the displayed asymptotic is incorrect and must be fixed.

minor comments (3)

[Proposition 1, after Eq. (35)] The sentence 'for f≡1, the last expression implies that ∑ ρ(u)^q > 0 for every N>0' is not justified. Positivity of the variance gives nonnegativity only of the weighted double sum, and the conclusion about the infinite sum is at best a limiting statement.
[Eqs. (57)–(59)] Absolute values on ρ should be introduced explicitly before bounding by Eq. (54); as written the step from ρ(j-ℓ)^q to |ρ(j-ℓ)|^q is implicit.
[Throughout] Typographical issues: 'forth moment theorem' should be 'fourth'; 'reminder' should be 'remainder'. The proof of Theorem 3 is labelled a sketch, but the convergence of finite-dimensional distributions of L_N + R_N to the claimed GFF is asserted rather than demonstrated.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: Theorem 1 is proved from the external Nualart–Peccati fourth-moment theorem and standard chaos isometries; the odd-power GFF gap is a scaling/correctness error rather than a circular reduction.

full rationale

Theorem 1's derivation is self-contained: the field is expanded into Hermite chaoses, S_{N,q}(f) is written as I_q(s_{N,q}), and Gaussianity is obtained from the Nualart–Peccati criterion (Theorem 4) plus a direct H^{-alpha} tightness estimate (Lemma 4). No parameter is fitted and no input is defined in terms of the white-noise covariance; the self-citations [CHRR23, CRS25] are contextual and do not carry the argument. The even-power GFF application genuinely follows from Theorem 1 under the stated summability (d>=5 for the field, d>=2 for the gradient). The real defect is in Section 1.3.2/Theorem 3: Eq. (19) asserts that c_1^2 N^{-d} sum f(i/N)f(j/N)G(i,j) converges to c_1^2 ∫∫ f(x)f(y)G_cont(x,y)dxdy, but for the discrete GFF G(i,j) ~ |i-j|^{2-d}, this variance is O(N^2) under that normalization, so Theorem 3 is mathematically unsupported. That is a correctness/scaling error, not circularity: the target covariance is asserted rather than derived, fitted, or defined in terms of the conclusion, and no self-citation supplies the missing step. Hence the circularity score is low; the paper's circularity is not the principal problem.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No free parameters are fitted and no new entities are introduced. The central mathematical machinery is standard Wiener chaos analysis. The main ad hoc input is the incorrect Riemann-sum scaling for the GFF odd-power application.

assumptions (4)

standard math Fourth-moment theorem of Nualart-Peccati characterizes Gaussianity of Wiener chaos elements.
Used as the main black box in Proposition 1; external theorem from [NP05].
domain assumption The discrete Gaussian free field on Z^d, d>=3, has covariance G(i,j) ~ |i-j|^{2-d}.
Standard property of the GFF, invoked in Sections 1.3.1 and 1.3.2.
ad hoc to paper N^{-d} sum_{i,j} f(i/N) f(j/N) G(i,j) converges to int int f G_cont g for the GFF Green's function.
This Riemann-sum assertion in the proof of Theorem 3 is false: the sum grows like N^2 for d>=3.
standard math Eigenfunction bound sup_{x,y} sum_k (1+lambda_k)^{-alpha} phi_k(x) phi_k(y) < infinity for alpha > d/2.
Used in the tightness proof, Lemma 4; the printed Weyl asymptotics lambda_k/k^2 -> 1 is false for d>1.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Central limit theorems for non-linear functionals of Gaussian fields via Wiener chaos decomposition." pith.science (2026). https://pith.science/paper/HZG3FNJJ

@misc{pith2026251213148,
  author       = {Pith},
  title        = {Pith review of: Central limit theorems for non-linear functionals of Gaussian fields via Wiener chaos decomposition},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HZG3FNJJ}},
  note         = {Machine review of arXiv:2512.13148}
}

abstract

We review and present some known results for non-linear functionals of Gaussian variables in the context of discrete Gaussian fields defined on the $d$ dimensional lattice. Our main result is a Central Limit Theorem in the spirit of the classical Breuer-Major theorem, together with applications to the powers of the Gaussian Free Field. Notably, we show that even powers of the discrete Gaussian Free Field converge to the Gaussian white noise, while odd powers converge to a continuous Gaussian Free Field with explicit covariance. The proofs are based on the Wiener chaos decomposition and the fourth moment theorem (Nualart-Peccati, 2005), and include a tightness result. Even if these tools are well-known in the literature, their application to Gaussian fields on the lattice appears to be new.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

37 extracted references · 5 linked inside Pith

[1]

Generalization of the Nualart – Peccati criterion

Ehsan Azmoodeh, Dominique Malicet, Guillaume Mijoule, and Guillaume Poly. Generalization of the Nualart – Peccati criterion. The Annals of Probability , 44(2), March 2016

2016
[2]

Miguel A. Arcones. Limit Theorems for Nonlinear Functionals of a Stationary Gaussian Sequence of Vectors . The Annals of Probability , 22(4):2242--2274, October 1994. Publisher: Institute of Mathematical Statistics

1994
[3]

Solesne Bourguin, Simon Campese, Nikolai Leonenko, and Murad S. Taqqu. Four moments theorems on Markov chaos. The Annals of Probability , 47(3), May 2019

2019
[4]

Central limit theorems for non-linear functionals of Gaussian fields

Péter Breuer and Péter Major. Central limit theorems for non-linear functionals of Gaussian fields. Journal of Multivariate Analysis , 13(3):425--441, September 1983

1983
[5]

Moment bounds and central limit theorems for Gaussian subordinated arrays

Jean-Marc Bardet and Donatas Surgailis. Moment bounds and central limit theorems for Gaussian subordinated arrays. Journal of Multivariate Analysis , 114:457--473, February 2013

2013
[6]

Fermionic Gaussian free field structure in the Abelian sandpile model and uniform spanning tree

Leandro Chiarini, Alessandra Cipriani, and Alan Rapoport. Fermionic Gaussian free field structure in the Abelian sandpile model and uniform spanning tree. 2024

2024
[7]

Hazra, Alan Rapoport, and Wioletta M

Alessandra Cipriani, Rajat S. Hazra, Alan Rapoport, and Wioletta M. Ruszel. Properties of the Gradient Squared of the Discrete Gaussian Free Field . Journal of Statistical Physics , 190(11):171, November 2023

2023
[8]

Continuous Breuer -major Theorem : Tightness and Nonstationarity

Simon Campese, Ivan Nourdin, and David Nualart. Continuous Breuer -major Theorem : Tightness and Nonstationarity . The Annals of Probability , 48(1):147--177, 2020. Publisher: Institute of Mathematical Statistics

2020

Show all 37 references

[9]

Ruszel, and Dirk Schuricht

Fabio Coppini, Wioletta M. Ruszel, and Dirk Schuricht. Wick theorem for analytic functions of Gaussian fields, July 2025. arXiv:2507.05131 [math]

2025
[10]

Central limit theorems for nonlinear functionals of stationary Gaussian processes

Daniel Chambers and Eric Slud. Central limit theorems for nonlinear functionals of stationary Gaussian processes. Probability Theory and Related Fields , 80(3):323--346, September 1989

1989
[11]

R. L. Dobrushin and P. Major. Non-central limit theorems for non-linear functional of Gaussian fields. Zeitschrift für Wahrscheinlichkeitstheorie und Verwandte Gebiete , 50(1):27--52, January 1979

1979
[12]

Fourth moment theorems on the Poisson space in any dimension

Christian Döbler, Anna Vidotto, and Guangqu Zheng. Fourth moment theorems on the Poisson space in any dimension. Electronic Journal of Probability , 23(none), January 2018. arXiv:1707.01889 [math]

2018 arXiv
[13]

The Fourth - Moment Theorem on Hilbert Spaces , September 2025

Marie-Christine Düker and Pavlos Zoubouloglou. The Fourth - Moment Theorem on Hilbert Spaces , September 2025. arXiv:2509.20342 [math]

2025
[14]

Spectral Criteria for the Asymptotics of Local Functionals of Gaussian Fields and Their Application to Nodal Volumes and Critical Points , June 2025

Louis Gass. Spectral Criteria for the Asymptotics of Local Functionals of Gaussian Fields and Their Application to Nodal Volumes and Critical Points , June 2025. arXiv:2501.07356 [math]

2025 arXiv
[15]

Equilibrium Fluctuations for grad-phi Interface Model

Giambattista Giacomin, Stefano Olla, and Herbert Spohn. Equilibrium Fluctuations for grad-phi Interface Model . The Annals of Probability , 29(3):1138--1172, 2001. Publisher: Institute of Mathematical Statistics

2001
[16]

Giraitis and D

L. Giraitis and D. Surgailis. CLT and other limit theorems for functionals of Gaussian processes. Zeitschrift für Wahrscheinlichkeitstheorie und verwandte Gebiete , 70(2):191--212, August 1985

1985
[17]

Renormalized self-intersection local time for fractional Brownian motion

Yaozhong Hu and David Nualart. Renormalized self-intersection local time for fractional Brownian motion. The Annals of Probability , 33(3), May 2005. arXiv:math/0506592

2005 arXiv
[18]

Gaussian Hilbert Spaces

Svante Janson. Gaussian Hilbert Spaces . Cambridge University Press, 1 edition, June 1997

1997
[19]

Wigner chaos and the fourth moment

Todd Kemp, Ivan Nourdin, Giovanni Peccati, and Roland Speicher. Wigner chaos and the fourth moment. The Annals of Probability , 40(4), July 2012

2012
[20]

Fluctuations for the Ginzburg - Landau \ \ \ nabla phi\ \ \ Interface Model on a Bounded Domain

Jason Miller. Fluctuations for the Ginzburg - Landau \ \ \ nabla phi\ \ \ Interface Model on a Bounded Domain . Communications in Mathematical Physics , 308(3):591--639, December 2011

2011
[21]

Limit theorems for the number of sign and level-set clusters of the Gaussian free field, January 2025

Michael McAuley and Stephen Muirhead. Limit theorems for the number of sign and level-set clusters of the Gaussian free field, January 2025. arXiv:2501.14707 [math]

2025 arXiv
[22]

Generalizations of the fourth moment theorem

Nobuaki Naganuma. Generalizations of the fourth moment theorem. Probability and Mathematical Statistics , 42(2), 2022

2022
[23]

C. M. Newman. Normal fluctuations and the FKG inequalities. Communications in Mathematical Physics , 74(2):119--128, January 1980. Publisher: Springer

1980
[24]

The functional Breuer – Major theorem

Ivan Nourdin and David Nualart. The functional Breuer – Major theorem. Probability Theory and Related Fields , 176(1):203--218, February 2020

2020
[25]

The Breuer – Major theorem in total variation: Improved rates under minimal regularity

Ivan Nourdin, David Nualart, and Giovanni Peccati. The Breuer – Major theorem in total variation: Improved rates under minimal regularity. Stochastic Processes and their Applications , 131:1--20, January 2021

2021
[26]

Lectures on Gaussian approximations with Malliavin calculus, June 2012

Ivan Nourdin. Lectures on Gaussian approximations with Malliavin calculus, June 2012. arXiv:1203.4147 [math]

2012 arXiv
[27]

Central limit theorems for sequences of multiple stochastic integrals

David Nualart and Giovanni Peccati. Central limit theorems for sequences of multiple stochastic integrals. The Annals of Probability , 33(1):177--193, January 2005. Publisher: Institute of Mathematical Statistics

2005
[28]

Stein’s method on Wiener chaos

Ivan Nourdin and Giovanni Peccati. Stein’s method on Wiener chaos. Probability Theory and Related Fields , 145(1):75--118, September 2009

2009
[29]

Quantitative Breuer – Major theorems

Ivan Nourdin, Peccati Peccati, and Mark Podolskij. Quantitative Breuer – Major theorems. Stochastic Processes and their Applications , 121(4):793--812, April 2011. Publisher: North-Holland

2011
[30]

Berry- Esseen bounds in the Breuer - Major CLT and Gebelein ’s inequality

Ivan Nourdin, Giovanni Peccati, and Xiaochuan Yang. Berry- Esseen bounds in the Breuer - Major CLT and Gebelein ’s inequality. Electronic Communications in Probability , 24(none), January 2019

2019
[31]

On homogenization and scaling limit of some gradient perturbations of a massless free field

Ali Naddaf and Thomas Spencer. On homogenization and scaling limit of some gradient perturbations of a massless free field. Communications in Mathematical Physics , 183(1):55--84, January 1997

1997
[32]

Continuous Breuer - Major theorem for vector valued fields

David Nualart and Abhishek Tilva. Continuous Breuer - Major theorem for vector valued fields. Stochastic Analysis and Applications , 38(4):668--685, July 2020

2020
[33]

Malliavin calculus and normal approximations

David Nualart. Malliavin calculus and normal approximations. Ensaios Matemáticos , 34(1), 2019

2019
[34]

Newman and Wei Wu

Charles M. Newman and Wei Wu. Gaussian fluctuations for the classical XY model. Annales de l'Institut Henri Poincaré, Probabilités et Statistiques , 54(4):1759--1777, November 2018. Publisher: Institut Henri Poincaré

2018
[35]

Giovanni Peccati and Ciprian A. Tudor. Gaussian Limits for Vector -valued Multiple Stochastic Integrals . In Michel Émery, Michel Ledoux, and Marc Yor, editors, Séminaire de Probabilités XXXVIII , pages 247--262. Springer, Berlin, Heidelberg, 2005

2005
[36]

Giovanni Peccati and Murad S. Taqqu. Wiener Chaos : Moments , Cumulants and Diagrams , volume 1 of Bocconi & Springer Series . Springer Milan, Milano, 2011

2011
[37]

Local central limit theorem for gradient field models

Wei Wu. Local central limit theorem for gradient field models. Probability Theory and Related Fields , 191(1):1--40, February 2025

2025

Pith tools

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

[1] [1]

Generalization of the Nualart – Peccati criterion

Ehsan Azmoodeh, Dominique Malicet, Guillaume Mijoule, and Guillaume Poly. Generalization of the Nualart – Peccati criterion. The Annals of Probability , 44(2), March 2016

2016

[2] [2]

Miguel A. Arcones. Limit Theorems for Nonlinear Functionals of a Stationary Gaussian Sequence of Vectors . The Annals of Probability , 22(4):2242--2274, October 1994. Publisher: Institute of Mathematical Statistics

1994

[3] [3]

Solesne Bourguin, Simon Campese, Nikolai Leonenko, and Murad S. Taqqu. Four moments theorems on Markov chaos. The Annals of Probability , 47(3), May 2019

2019

[4] [4]

Central limit theorems for non-linear functionals of Gaussian fields

Péter Breuer and Péter Major. Central limit theorems for non-linear functionals of Gaussian fields. Journal of Multivariate Analysis , 13(3):425--441, September 1983

1983

[5] [5]

Moment bounds and central limit theorems for Gaussian subordinated arrays

Jean-Marc Bardet and Donatas Surgailis. Moment bounds and central limit theorems for Gaussian subordinated arrays. Journal of Multivariate Analysis , 114:457--473, February 2013

2013

[6] [6]

Fermionic Gaussian free field structure in the Abelian sandpile model and uniform spanning tree

Leandro Chiarini, Alessandra Cipriani, and Alan Rapoport. Fermionic Gaussian free field structure in the Abelian sandpile model and uniform spanning tree. 2024

2024

[7] [7]

Hazra, Alan Rapoport, and Wioletta M

Alessandra Cipriani, Rajat S. Hazra, Alan Rapoport, and Wioletta M. Ruszel. Properties of the Gradient Squared of the Discrete Gaussian Free Field . Journal of Statistical Physics , 190(11):171, November 2023

2023

[8] [8]

Continuous Breuer -major Theorem : Tightness and Nonstationarity

Simon Campese, Ivan Nourdin, and David Nualart. Continuous Breuer -major Theorem : Tightness and Nonstationarity . The Annals of Probability , 48(1):147--177, 2020. Publisher: Institute of Mathematical Statistics

2020