Quenched and Annealed CLTs for the one-periodic Aztec diamond in random environment

Panagiotis Zografos

REVIEW 3 major objections 4 minor 1 cited by

Quenched and Annealed CLTs for the one-periodic Aztec diamond in random environment

T0 review · 3 major / 4 minor · reviewed 2026-08-04 · deepseek-v4-flash

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

Pith's one-line read Random edge weights do not alter the quenched Gaussian Free Field fluctuations of the one-periodic Aztec diamond's height function.

desk verdict Quenched GFF universality for one-periodic random dimers is likely right, but the keystone concentration bound is only sketched. read the letter →

arxiv 2510.11846 v2 pith:TZ4G25KU submitted 2025-10-13 math.PR math-phmath.MP

Panagiotis Zografos This is my paper · ORCID

classification math.PRmath-phmath.MP MSC 60F0560K3582B20

keywords quenchedCLTannealedAztecdiamondrandomenvironmentGaussianFreeFieldSchurprocessheightfunctionedgeweights

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 proves that for the one-periodic Aztec diamond with random edge weights, conditioned on a typical realization of the disorder, the height function fluctuates exactly as it does for deterministic weights: the fluctuations converge to the Gaussian Free Field. This holds for a broad class of random environments, called CLT appropriate weights, with fluctuation scaling epsilon = 1/2 or 1, and without assuming independence of the weights. The same machinery establishes annealed central limit theorems under two different scalings: weights with classical sqrt(M) fluctuations give annealed height fluctuations that are Gaussian, typically a sum of two Brownian motions; weights such as GUE eigenvalues, whose fluctuations scale with M, give a Gaussian Free Field plus an independent, weaker Gaussian field. The approach works through the Schur-process representation of the tiling measure, using differential operators and contour integrals rather than Schur generating functions, because the latter cannot be simplified for non-independent random weights.

What carries the argument

The engine is the Schur-process representation of the one-periodic Aztec diamond measure, combined with differential operators D_k applied to the Cauchy identity. Repeated application yields moments of the height function as linear combinations of contour integrals (Lemmas 1-4); Lemma 4 decomposes f^{-1}D_{k_nu}...D_{k_1}f into products of F_k terms (limit shape), pairings of G_{k,l} terms (covariance), and a lower-order error term H. For the quenched CLT, Proposition 4 bounds the fourth (epsilon = 1/2) or second (epsilon = 1) moment of the quenched centered moments by order M^{-2}, giving a.s. convergence through a standard summability argument. The 'LLN appropriate' and 'epsilon-CLT approp

What would settle it

Compute explicitly the fourth-moment concentration bound in Proposition 4 for the simplest i.i.d. environment (e.g., binary weights at epsilon = 1/2), tracking every non-leading contour integral from Lemma 4; a non-vanishing term of order M^{-2} would break the a.s. convergence step. Or simulate the quenched covariance for large M and compare empirical finite-dimensional distributions to the Gaussian Free Field covariance (44).

Watch

Extended reading notes

Core claim

Theorem 10 is the central claim: for epsilon-CLT appropriate weights (epsilon = 1/2 or 1), almost surely over the disorder the rescaled height-function fluctuations {M^{-k}(p_k - E_lambda[p_k])} converge in finite-dimensional distribution to a Gaussian vector with the Gaussian Free Field covariance (44), the same as for deterministic weights; environmental fluctuations drop out entirely. Theorems 8 and 9 give annealed CLTs: on the sqrt(M) scale (epsilon = 1/2) the annealed fluctuations are Gaussian; on the M scale (epsilon = 1) they are the Gaussian Free Field plus an independent, weaker Gaussian field. Proposition 4 provides quenched concentration that upgrades the annealed convergence to a

Load-bearing premise

The quenched CLT relies on Proposition 4, whose proof identifies the leading term of the quenched moments and asserts that all remaining contour-integral terms cancel or vanish without presenting the cancellations; if those cancellations fail, the almost-sure convergence to the Gaussian Free Field covariance is unsupported.

Editorial extensions

If this is right

Almost surely in the disorder, the height function of the one-periodic Aztec diamond has Gaussian Free Field fluctuations, exactly as in the deterministic model (Theorem 10).
The annealed fluctuations of the height function for epsilon = 1/2 weights occur on scale sqrt(M) and are Gaussian; for i.i.d. or Markov-chain weights they are a sum of two correlated Brownian motions (Corollary 1).
For epsilon = 1 weights (e.g., GUE eigenvalues), annealed fluctuations occur on scale M and are the Gaussian Free Field plus an independent Gaussian field with weaker singularity (Corollary 2).
A law of large numbers holds for the height function under non-independent, LLN-appropriate weights (Theorem 3), giving explicit limit-shape formulas.
The absence of independence in the environment is handled: the method does not require the Schur generating function to simplify (Section 3).

Reading between the lines

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

Because the quenched covariance (44) depends only on the limit shapes of the empirical measures of the weights (through F1 and F2), the quenched GFF universality should extend to any environment with the same limiting empirical measures, even with long-range correlations.
The epsilon = 1/2 versus epsilon = 1 dichotomy suggests a threshold: environmental fluctuations of order M^{1/2} are invisible in the quenched limit, while fluctuations of order M contribute a separate Gaussian field; studying epsilon in (1/2,1) with intermediate correlations could interpolate between the two regimes.
Proposition 4's proof is a sketch that identifies the leading term and asserts cancellations; a fully expanded verification of those cancellations would either confirm the quenched CLT as stated or reveal that the covariance (44) must be corrected for certain environments.
The contour-integral framework may transfer to other random Schur processes, such as lozenge tilings with random weights or higher-periodic Aztec diamonds, where Schur generating functions also fail to simplify.

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 headline: this paper gives a real extension of the Bufetov–Petrov–Zografos program, and the quenched CLT is the genuinely new piece. But the proof of the quenched result rests on a concentration estimate, Proposition 4, that is only sketched. That is the soft spot you need to know about.

What is actually new: the quenched CLT (Theorem 10) for one-periodic Aztec diamonds with CLT-appropriate random weights, plus annealed CLTs for non-independent weights (Markov chains, GUE eigenvalues). The unified contour-integral treatment of the two scalings (ε=1/2 and ε=1) is coherent, and the explicit GUE example is a nice concrete payoff. The limit-shape formula genuinely generalizes Proposition 5.12 of [15]. The paper is also honest about what it defers.

Where the softness is real: Proposition 4 is load-bearing. For ε=1/2 it requires M^2 E[(X−EX)^4] to converge; for ε=1, M^2 E[(X−EX)^2]. Theorem 10 feeds those bounds into Borel–Cantelli to get almost-sure convergence. The proof identifies the candidate leading term (46), then says the other terms “either will cancel out or vanish,” citing arguments like Theorem 5. That is not carried out. The concern in the stress-test note is fair: the quenched setting is different because the outer average is over the environment and the centered quantity is already quenched, so the cancellation mechanism is not obviously the same. The same pattern appears in Lemma 3(2), which is explicitly omitted, and in Theorem 7, whose cancellation analysis is summarized. These are not fatal in themselves—the architecture is plausible and the annealed case is worked out in much more detail—but they are exactly where a referee needs to dig.

Bottom line: if Proposition 4 holds, this is a solid contribution. Right now it is a strong conditional accept. It deserves a serious referee, and I would send it to review with instructions to focus on Proposition 4 and on the GUE covariance computation. For my own reading group, I would probably wait until the missing steps are supplied; the method is interesting but the main theorem is not yet verifiable.

Referee Report

3 major / 4 minor

Summary. The paper studies the one-periodic Aztec diamond with random edge weights that are assumed to satisfy a moment-type CLT condition (Definition 2), without requiring independence. The main results are: (i) an annealed law of large numbers for the height function (Theorem 3); (ii) annealed CLTs in two scaling regimes, ε=1/2 (Theorem 5, multilevel Theorem 8) and ε=1 (Theorem 7, multilevel Theorem 9), with explicit covariances; and (iii) a quenched CLT (Theorem 10) asserting that, almost surely in the disorder, centered and rescaled height-function moments converge to the Gaussian Free Field covariance (44). The proofs are based on Schur-process contour-integral expansions, in particular a decomposition of f^{-1}D_{k_ν}...D_{k_1}f into products of F_k, G_{k,l}, and higher-order contour integrals. The paper is an extension of the author's earlier work with Bufetov and Petrov [15] to non-i.i.d. CLT-appropriate weights.

Significance. If the main theorems hold, the paper makes a meaningful contribution to the random-environment dimer literature: it shows that the quenched fluctuations remain in the Gaussian Free Field universality class, while annealed fluctuations are dominated by the environment randomness on a larger scale. The explicit contour-integral covariance formulas are a strength, and the framework avoids Schur generating functions, covering examples such as Markov chains and GUE eigenvalues. The paper is also honest in that it introduces no fitted constants: the functions F_1,F_2 and G_1,G_2,G_3 are limits of moments of the given environment. However, the quenched CLT rests on Proposition 4, whose proof is only a sketch, and a number of cancellation analyses are delegated; the significance is therefore conditional on closing those gaps.

major comments (3)

[Sec. 6, Proposition 4] This is the load-bearing concentration estimate for the quenched CLT. The proof identifies the would-be leading term (46) and then states that all other terms 'either will cancel out or they will vanish as M→∞', referring to 'similar arguments as in Theorem 5'. That is not a proof. In Theorem 5 and Proposition 3 the cancellations are nontrivial: they rely on the binomial identities in (26) and on the pairing structure of the stochastic factors (η,θ) in the F_k expansion. The quenched X_{k_1,...,k_r} is already an environment-dependent expectation, and the higher-order terms are products of the G^{(N_i,N_j)}_{k_i,k_j} objects, so the cancellation mechanism is not obviously the same. Without a complete proof of Proposition 4, the bound needed for the Borel–Cantelli step at (49) is unsupported, and the almost-sure convergence in (48) does not follow.
[Sec. 3.2, Lemma 3(2)] Part 2 of Lemma 3 is stated without proof: 'The proof of 2 is similar and it is omitted.' This lemma is used in the induction proving Lemma 4, and Lemma 4 underpins essentially every moment expansion in Sections 4–6 (Theorems 4, 5, 7, 8, 9, 10). A statement of this importance cannot be left as an omitted similar argument. Please supply the proof in full or give a precise reference to a written proof.
[Sec. 4.2.2, Theorem 7] The proof of the 1-CLT annealed theorem is summarized at the critical point. After deriving the leading contribution (35) and the pair-sum expression (36), the proof says that the remaining terms 'do not contribute to the limit' for 'very similar reasons as in Theorem 5 and Proposition 3'. This is more delicate in the ε=1 regime than in the ε=1/2 regime because the G_{k,l} terms themselves contribute to the limiting covariance, and the bookkeeping of the (η,θ) tuples must account for both F_k and G_{k,l} factors. As written, the argument is not checkable. Since Theorem 7 is one of the main annealed results, the omitted cancellation analysis is a load-bearing gap.

minor comments (4)

[Sec. 5, Example 4] After equation (42), the functions G_1(z,w) and G_2(z,w) are not written out because the formulas are 'too long and omitted'. This example is used to support Corollary 2 and the claim that the covariance has singularity weaker than the GFF. Since the computation is omitted, the reader cannot verify the example. At minimum, state the singular behavior or provide the computation in supplementary material.
[Sec. 6, Theorem 10] The almost-sure statement in Theorem 10 involves the uncountable family of parameters 0<γ≤1 and all finite tuples (γ_1,...,γ_r). The proof applies Borel–Cantelli to a fixed tuple (k_1,...,k_r; γ_1,...,γ_r) and does not explain how to pass to a simultaneous almost-sure statement over all γ. If the theorem is only claimed for each fixed tuple, the quantifier should be stated precisely; if the stronger simultaneous statement is intended, a countable-reduction or regularity argument is needed.
[Sec. 2, Convention 1] The support condition says x_i is supported on (δ,2−δ) for some δ∈(0,2). This interval is empty for δ≥1. The convention should state δ∈(0,1) or otherwise ensure (δ,2−δ) is nonempty.
[Sec. 6, Theorem 10] In the second bullet, 'p^{(N1.M)}_l' contains a typo: the dot should be a comma (N_1,M).

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the quenched CLT is derived from Schur-process moment expansions; the main fragility is an unproved cancellation estimate, not a circular reduction.

full rationale

The derivation chain is not circular. The quenched CLT (Theorem 10) rests on Proposition 4, whose proof is explicitly a sketch: it states that 'All the other terms ... either will cancel out or they will vanish as M→∞' (Section 6, proof of Proposition 4) and refers to 'similar arguments as in Theorem 5'. This is a genuine rigor gap that makes the almost-sure conclusion conditional, but it is an omitted proof, not an equivalence between input and output. The covariance (44) is obtained from contour-integral expansions of Schur-process moments (Lemmas 1-4), not from assuming the GFF; the (z-w)^{-2} term emerges from a double-contour integral and is identified with the GFF only afterwards. The functions F1, F2, G1-G3 are limits of moments of the given random environment, not fitted constants, and no parameter is calibrated to the target height fluctuations. Self-citations to [15] appear (e.g., Example 1 recovers Proposition 5.12 of [15], and the text says the results generalize [15]), but they are used for context and generalization statements, not as load-bearing evidence for Theorem 10. No uniqueness theorem or ansatz is imported from the authors' prior work. Hence no circular step is present.

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

The paper rests on standard Schur/contour machinery plus two domain assumptions: Convention 1's support condition and Definition 2's CLT-appropriateness. No free parameters are fitted; F1,F2,G1-G3 are limits of moments of the given environment, not adjustable constants. No invented entities are introduced.

assumptions (5)

domain assumption The one-periodic Aztec diamond edge-weight model is equivalent to the Schur process of Proposition 1 (measure P_{β,y}).
Section 2.1, Proposition 1, cited from [24],[14]. The entire proof uses this representation.
domain assumption Convention 1: β_i is supported in (0,1) and x_i in (δ,2−δ), so the power series in (17)-(18) converge uniformly on the chosen contours.
Section 2.1, Convention 1. This support/analyticity condition is needed for the residue expansions that drive the asymptotics; unbounded or boundary-supported weights require modifications (Remark 1).
domain assumption Definition 2: the random weights are ε-CLT appropriate, meaning their partial sums of powers converge in the sense of moments to a Gaussian vector and the relevant covariance power series converge uniformly.
Definition 2, Section 4. This is the main modeling assumption; both quenched and annealed CLTs are conditional on it and it is not proved for arbitrary examples.
standard math Standard Schur function identities: determinantal formula, the dual Cauchy identity (5), and the summation formula (3).
Sections 2.2 and 3; classical results from [10],[24],[31],[32].
standard math Contour integral and residue theorem manipulations underlying Lemmas 1-4.
Section 3; standard complex analysis used to express D_k derivatives as contour integrals and to extract leading-order terms.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Quenched and Annealed CLTs for the one-periodic Aztec diamond in random environment." pith.science (2026). https://pith.science/paper/TZ4G25KU

@misc{pith2026251011846,
  author       = {Pith},
  title        = {Pith review of: Quenched and Annealed CLTs for the one-periodic Aztec diamond in random environment},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TZ4G25KU}},
  note         = {Machine review of arXiv:2510.11846}
}

read the original abstract

We study the asymptotic behavior of random dimer coverings of the one-periodic Aztec diamond in random environment. We investigate quenched limit theorems for the height function and we extend annealed limit theorems that were recently studied in [arXiv:2507.08560]. We consider more general choices of random edge weights (independence is not assumed) and we distinguish two cases where the random edge weights satisfy the Central Limit Theorem (CLT) under different scalings. For both cases, we prove convergence to the Gaussian Free Field for the quenched fluctuations. For the annealed version, it had been shown in [arXiv:2507.08560], that Gaussian Free Field fluctuations can be dominated by the much larger fluctuations of the random environment. To access quenched fluctuations we analyze the Schur process with random parameters in a way that allows to prove the annealed CLT for the height function for non i.i.d. weights. We consider specific examples where we determine the asymptotic fluctuations.

Figures

Figures reproduced from arXiv: 2510.11846 by the authors.

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

Non-colliding space-time inhomogeneous Markov chains
math.PR 2026-06 unverdicted novelty 7.0 of 10

Derives leading asymptotics for collision-time tails of integrable inhomogeneous Markov chains via steepest-descent analysis and Karlin-McGregor expansion, confirming a prediction for push-block particle systems.

Reference graph

Works this paper leans on

36 extracted references · 8 linked inside Pith · cited by 1 Pith paper

[15]

Domino Tilings of the Aztec Diamond in Random Environment and Schur Generating Functions.arXiv preprint, arXiv:2507.08560, 2025

Alexey Bufetov, Leonid Petrov, and Panagiotis Zografos. Domino Tilings of the Aztec Diamond in Random Environment and Schur Generating Functions.arXiv preprint, arXiv:2507.08560, 2025

arXiv 2025
[1]

Central limit theorem for partial linear eigenvalue statistics of Wigner matrices.J

Zhigang Bao, Guangming Pan, and Wang Zhou. Central limit theorem for partial linear eigenvalue statistics of Wigner matrices.J. Stat. Phys., 150(1):88–129, 2013

2013
[2]

Dimers and imaginary geometry.Ann

Nathana¨ el Berestycki, Beno ˆ ıt Laslier, and Gourab Ray. Dimers and imaginary geometry.Ann. Probab., 48(1):1–52, 2020

2020
[3]

Domino tilings of the Aztec diamond with doubly periodic weightings

Tomas Berggren. Domino tilings of the Aztec diamond with doubly periodic weightings. Ann. Probab., 49(4):1965–2011, 2021

1965
[4]

Crystallization of the Aztec diamond.arXiv preprint, arXiv:2410.04187, 2024

Tomas Berggren and Alexei Borodin. Crystallization of the Aztec diamond.arXiv preprint, arXiv:2410.04187, 2024

arXiv 2024
[5]

Geometry of the doubly periodic Aztec dimer model

Tomas Berggren and Alexei Borodin. Geometry of the doubly periodic Aztec dimer model. Commun. Am. Math. Soc., 5:475–570, 2025

2025
[6]

Gaussian Free Field and Discrete Gaussians in Periodic Dimer Models.arXiv preprint, arXiv:2502.07241, 2025

Tomas Berggren and Matthew Nicoletti. Gaussian Free Field and Discrete Gaussians in Periodic Dimer Models.arXiv preprint, arXiv:2502.07241, 2025. 33

arXiv 2025
[7]

Bobenko and Nikolai Bobenko

Alexander I. Bobenko and Nikolai Bobenko. Dimers and M-curves: Limit shapes from Riemann surfaces.arXiv preprint, arXiv:2407.19462, 2024

arXiv 2024

Show all 36 references

[8]

Alexei Borodin and Patrik L. Ferrari. Anisotropic growth of random surfaces in 2 + 1 dimensions. Comm. Math. Phys., 325(2):603–684, 2014

2014
[9]

Fock’s dimer model on the Aztec diamond.arXiv preprint, arXiv:2405.20284, 2024

C´ edric Boutillier and B´ eatrice de Tili` ere. Fock’s dimer model on the Aztec diamond.arXiv preprint, arXiv:2405.20284, 2024

2024 arXiv
[10]

Limit shape and height fluctuations of random perfect matchings on square-hexagon lattices.Ann

C´ edric Boutillier and Zhongyang Li. Limit shape and height fluctuations of random perfect matchings on square-hexagon lattices.Ann. Inst. Fourier (Grenoble), 71(6):2305–2386, 2021

2021
[11]

Representations of classical Lie groups and quantized free convo- lution.Geom

Alexey Bufetov and Vadim Gorin. Representations of classical Lie groups and quantized free convo- lution.Geom. Funct. Anal., 25(3):763–814, 2015

2015
[12]

Fluctuations of particle systems determined by Schur generating functions.Adv

Alexey Bufetov and Vadim Gorin. Fluctuations of particle systems determined by Schur generating functions.Adv. Math., 338:702–781, 2018

2018
[13]

Fourier transform on high-dimensional unitary groups with ap- plications to random tilings.Duke Math

Alexey Bufetov and Vadim Gorin. Fourier transform on high-dimensional unitary groups with ap- plications to random tilings.Duke Math. J., 168(13):2559–2649, 2019

2019
[14]

Asymptotics of random domino tilings of rectangular Aztec dia- monds.Ann

Alexey Bufetov and Alisa Knizel. Asymptotics of random domino tilings of rectangular Aztec dia- monds.Ann. Inst. Henri Poincar´ e Probab. Stat., 54(3):1250–1290, 2018

2018
[16]

Domino statistics of the two-periodic Aztec diamond.Adv

Sunil Chhita and Kurt Johansson. Domino statistics of the two-periodic Aztec diamond.Adv. Math., 294:37–149, 2016

2016
[17]

Asymptotic domino statistics in the Aztec diamond.Ann

Sunil Chhita, Kurt Johansson, and Benjamin Young. Asymptotic domino statistics in the Aztec diamond.Ann. Appl. Probab., 25(3):1232–1278, 2015

2015
[18]

Coupling functions for domino tilings of Aztec diamonds

Sunil Chhita and Benjamin Young. Coupling functions for domino tilings of Aztec diamonds. Adv. Math., 259:173–251, 2014

2014
[19]

Maurice Duits and Arno B. J. Kuijlaars. The two-periodic Aztec diamond and matrix valued or- thogonal polynomials.J. Eur. Math. Soc. (JEMS), 23(4):1075–1131, 2021

2021
[20]

Alternating-sign matrices and domino tilings

Noam Elkies, Greg Kuperberg, Michael Larsen, and James Propp. Alternating-sign matrices and domino tilings. I.J. Algebraic Combin., 1(2):111–132, 1992

1992
[21]

Cambridge University Press, Cambridge, 2021

Vadim Gorin.Lectures on Random Lozenge Tilings, volume 193 ofCambridge Studies in Advanced Mathematics. Cambridge University Press, Cambridge, 2021

2021
[22]

Random Domino Tilings and the Arctic Circle theorem.arXiv preprint, arXiv:math/9801068, 1998

William Jockusch, James Propp, and Peter Shor. Random Domino Tilings and the Arctic Circle theorem.arXiv preprint, arXiv:math/9801068, 1998

1998 arXiv
[23]

On fluctuations of eigenvalues of random Hermitian matrices.Duke Math

Kurt Johansson. On fluctuations of eigenvalues of random Hermitian matrices.Duke Math. J., 91(1):151–204, 1998

1998
[24]

Non-intersecting paths, random tilings and random matrices.Probab

Kurt Johansson. Non-intersecting paths, random tilings and random matrices.Probab. Theory Relat. Fields, 123(2):225–280, 2002

2002
[25]

The arctic circle boundary and the Airy process.Ann

Kurt Johansson. The arctic circle boundary and the Airy process.Ann. Probab., 33(1):1–30, 2005

2005
[26]

Dominos and the Gaussian free field.Ann

Richard Kenyon. Dominos and the Gaussian free field.Ann. Probab., 29(3):1128–1137, 2001. 34

2001
[27]

Height fluctuations in the honeycomb dimer model.Comm

Richard Kenyon. Height fluctuations in the honeycomb dimer model.Comm. Math. Phys., 281(3):675–709, 2008

2008
[28]

Dimers and amoebae.Ann

Richard Kenyon, Andrei Okounkov, and Scott Sheffield. Dimers and amoebae.Ann. of Math. (2), 163(3):1019–1056, 2006

2006
[29]

Two-periodic weighted dominos and the sine-Gordon field at the free fermion point: I.arXiv preprint, arXiv:2209.11111, 2023

Scott Mason. Two-periodic weighted dominos and the sine-Gordon field at the free fermion point: I.arXiv preprint, arXiv:2209.11111, 2023

2023 arXiv
[30]

Dimers with layered disorder.arXiv preprint, arXiv:2507.11964, 2025

Quentin Moulard and Fabio Toninelli. Dimers with layered disorder.arXiv preprint, arXiv:2507.11964, 2025

2025 arXiv
[31]

Infinite wedge and random partitions.Selecta Math

Andrei Okounkov. Infinite wedge and random partitions.Selecta Math. (N.S.), 7(1):57–81, 2001

2001
[32]

Correlation function of Schur process with application to local geometry of a random 3-dimensional Young diagram.J

Andrei Okounkov and Nikolai Reshetikhin. Correlation function of Schur process with application to local geometry of a random 3-dimensional Young diagram.J. Amer. Math. Soc., 16(3):581–603, 2003

2003
[33]

Anthony Perret, Zoran Ristivojevic, Pierre Le Doussal, Gr´ egory Schehr, and Kay J. Wiese. Super- Rough Glassy Phase of the Random FieldXYModel in Two Dimensions.Phys. Rev. Lett., 109:157205, Oct 2012

2012
[34]

Asymptotics of uniformly random lozenge tilings of polygons

Leonid Petrov. Asymptotics of uniformly random lozenge tilings of polygons. Gaussian free field. Ann. Probab., 43(1):1–43, 2015

2015
[35]

Convergence of moments in a Markov-chain central limit theorem.Indag

David Steinsaltz. Convergence of moments in a Markov-chain central limit theorem.Indag. Math. (N.S.), 12(4):533–555, 2001

2001
[36]

Thurston

William P. Thurston. Conway’s tiling groups.Amer. Math. Monthly, 97(8):757–773, 1990. (Panagiotis Zografos) Institute of Mathematics, Leipzig University, Augustusplatz 10, 04109 Leipzig, Germany Email address: pzografos04@gmail.com 35

1990

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.

[1] [15]

Domino Tilings of the Aztec Diamond in Random Environment and Schur Generating Functions.arXiv preprint, arXiv:2507.08560, 2025

Alexey Bufetov, Leonid Petrov, and Panagiotis Zografos. Domino Tilings of the Aztec Diamond in Random Environment and Schur Generating Functions.arXiv preprint, arXiv:2507.08560, 2025

arXiv 2025

[2] [1]

Central limit theorem for partial linear eigenvalue statistics of Wigner matrices.J

Zhigang Bao, Guangming Pan, and Wang Zhou. Central limit theorem for partial linear eigenvalue statistics of Wigner matrices.J. Stat. Phys., 150(1):88–129, 2013

2013

[3] [2]

Dimers and imaginary geometry.Ann

Nathana¨ el Berestycki, Beno ˆ ıt Laslier, and Gourab Ray. Dimers and imaginary geometry.Ann. Probab., 48(1):1–52, 2020

2020

[4] [3]

Domino tilings of the Aztec diamond with doubly periodic weightings

Tomas Berggren. Domino tilings of the Aztec diamond with doubly periodic weightings. Ann. Probab., 49(4):1965–2011, 2021

1965

[5] [4]

Crystallization of the Aztec diamond.arXiv preprint, arXiv:2410.04187, 2024

Tomas Berggren and Alexei Borodin. Crystallization of the Aztec diamond.arXiv preprint, arXiv:2410.04187, 2024

arXiv 2024

[6] [5]

Geometry of the doubly periodic Aztec dimer model

Tomas Berggren and Alexei Borodin. Geometry of the doubly periodic Aztec dimer model. Commun. Am. Math. Soc., 5:475–570, 2025

2025

[7] [6]

Gaussian Free Field and Discrete Gaussians in Periodic Dimer Models.arXiv preprint, arXiv:2502.07241, 2025

Tomas Berggren and Matthew Nicoletti. Gaussian Free Field and Discrete Gaussians in Periodic Dimer Models.arXiv preprint, arXiv:2502.07241, 2025. 33

arXiv 2025

[8] [7]

Bobenko and Nikolai Bobenko

Alexander I. Bobenko and Nikolai Bobenko. Dimers and M-curves: Limit shapes from Riemann surfaces.arXiv preprint, arXiv:2407.19462, 2024

arXiv 2024