Domino Tilings of the Aztec Diamond in Random Environment and Schur Generating Functions

Alexey Bufetov; Leonid Petrov; Panagiotis Zografos

REVIEW 4 major objections 6 minor 2 cited by

Domino Tilings of the Aztec Diamond in Random Environment and Schur Generating Functions

T0 review · 4 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read Random one-periodic edge weights on the Aztec diamond add an independent Brownian motion to the Gaussian Free Field when their variance decays as 1/M; fixed weight distributions instead give larger sqrt(M) fluctuations described by…

desk verdict A genuinely new random-environment dimer model and a substantial extension of the Schur generating function method, but the advertised GFF-plus-Brownian height-function limit is stronger than what the proofs actually support. read the letter →

arxiv 2507.08560 v1 pith:RLU6L6KI submitted 2025-07-11 math.PR cond-mat.stat-mechmath-phmath.COmath.MP

Alexey Bufetov , Leonid Petrov , Panagiotis Zografos This is my paper · ORCID

classification math.PRcond-mat.stat-mechmath-phmath.COmath.MP

keywords AztecdiamonddominotilingsrandomedgeweightsSchurgeneratingfunctionsGaussianFreeFieldBrownianmotioncentrallimittheoremenvironment

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 studies domino tilings of the Aztec diamond whose edge weights are random: a one-periodic pattern of i.i.d. weights, meaning the weights vary along one lattice direction and repeat along the other. It tries to establish how this random environment changes the fluctuations of the tiling's height function compared with the uniform, deterministic-weight case. In the critical regime, where the weight variance is of order $1/M$ for an $M \times M$ diamond, the paper claims the unrescaled fluctuations split into two independent parts: the Gaussian Free Field already present for fixed weights, plus a one-dimensional Brownian motion coming from the random environment. In the fixed-distribution regime, it claims the fluctuations occur at the much larger scale $\sqrt{M}$ and are governed by Brownian motion alone, with a new limit shape generally different from any fixed-weight tiling. The proofs go through Schur generating functions, and the paper's generalized law of large numbers and central limit theorem for such particle systems are stated for their own sake as well.

What carries the argument

The key object is the Schur generating function of a probability measure on signatures, $\sum_\lambda \rho_N(\lambda) s_\lambda(x_1,\dots,x_N)/s_\lambda(1^N)$, together with differential operators that act diagonally on Schur functions. For an i.i.d. environment the annealed Schur generating function factorizes into the simple product $(\mathbb{E}_B \prod_{i=1}^k (1-\beta+x_i\beta))^{M-N}$, and the paper shows that the $N$-th root of this function determines the law of large numbers and the central limit theorem. The proofs rest on novel symmetrization identities that evaluate derivatives at points scaled by roots of unity, converting the factorized generating function into the explicit moment and covariance formulas used for the Aztec diamond.

What would settle it

Pick a distribution $B$ with known moments, compute the multilevel covariance from formula (4.5) for three slice levels $\alpha_1<\alpha_2<\alpha_3$, and check whether the weight-dependent term satisfies the Brownian increment relation $\mathrm{Cov}(\alpha_1,\alpha_3)=\mathrm{Cov}(\alpha_1,\alpha_2)+\mathrm{Cov}(\alpha_2,\alpha_3)$ after mapping the liquid region to the upper half-plane; if the relation fails for any triple, the Brownian-motion interpretation is wrong, and the moment-level theorems would not imply the informal height-function claims.

Watch

Extended reading notes

Core claim

The central discovery is that a random environment adds a new, universal fluctuation component to Aztec diamond tilings, and its size is controlled by how the weight variance scales with $M$. When $\mathrm{Var}(B_M) \sim \sigma^2/M$, the annealed height function keeps the same deterministic limit shape as constant weights $\beta$, but its unrescaled fluctuations converge at the level of moments to the sum of the Gaussian Free Field and an independent Brownian motion, with the covariance decomposition made explicit in Theorems 4.4 and 4.5. When the weight distribution is fixed, Theorems 5.6 and 6.2 supply a law of large numbers and a central limit theorem in which the normalized moments have Gaussian fluctuations at scale $N^{k+1/2}$, and Theorem 7.1 identifies the limiting covariance directly as a Brownian-motion-type covariance with kernel $\mathrm{Cov}_B(\beta/(1-\beta+\beta z), \beta/(1-\beta+\beta w))$. The informal Theorems 1.1 and 1.2 state the height-function-level version of these moment results.

Load-bearing premise

The load-bearing assumption is that the covariance term coming from the random weights really is a Brownian motion after the coordinate change to the upper half-plane; the paper verifies this only informally and proves the theorems at the level of moments, so if the coordinate map or the identification fails the claimed height-function description would not follow.

Editorial extensions

If this is right

In the critical regime, the Gaussian Free Field remains visible in the limit even though the environment is random, so small random perturbations of edge weights do not destroy the uniform tiling's fluctuation structure; they add an independent Brownian layer on top.
In the fixed-variance regime, the height fluctuations grow at $\sqrt{M}$, the same scale as the fluctuations of the empirical mean of $M$ i.i.d. weights, so the environment's own randomness dominates the tiling's internal fluctuations.
The paper's limit shape for fixed random weights is generally not the ellipse of the uniform Aztec diamond; for discrete distributions it has an exact arctic-curve parametrization via double roots of the equation $F(z;\alpha^{-1}-1)=\alpha^{-1}y$.
The generalized law of large numbers and central limit theorem for Schur generating functions apply to any sequence of measures on signatures whose $N$-th-root generating functions converge, not only to the domino tilings treated here.

Reading between the lines

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

The paper leaves implicit that the Brownian component in the critical regime is a universal function of the variance $\sigma^2$ and not of the shape of the weight distribution; one could test this by holding $\sigma^2$ fixed while varying the distribution $B$ and checking that the limiting covariance is unchanged.
Because the factorization of the annealed generating function only uses the independence of the environment, the same Schur-generating-function approach should extend to two-periodic or short-range-correlated random weights; a concrete next step is deriving the analog of the factorization for correlated $\beta_i$ and checking whether a Brownian term still appears.
In the fixed-variance regime, the dominance of the environment's fluctuations suggests a quenched-annealed equivalence: for almost every realization of the weights, the height profile should converge to the same annealed limit shape, with fluctuations governed by the empirical process of the weights.

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

You should read this for the fixed-variance regime: random one-periodic weights on the Aztec diamond, fluctuations on scale sqrt M, Brownian motion only. That is a real new phenomenon, not present in the deterministic-weight literature, and it is outside what [BG15]/[BG18] could reach. The Schur generating function route is the right one, and Proposition 3.2 — reducing the annealed generating function to a product over i.i.d. betas — is clean and genuinely useful.

What the paper actually proves, rigorously, is a pair of general limit theorems (Theorem 5.6 and Theorem 6.2) for particle systems defined by Schur generating functions, plus their application to random weights. Those theorems are the main new technical content: the N-th root formulation, the root-of-unity moment formulas, and the fluctuation scale N^{k+1/2} are all beyond the earlier framework. I believe those proofs are plausible and largely sound, though Section 6.2 is explicitly a sketch and Section 6.4 contains a corrupted, unreadable formula. That corruption needs fixing before publication.

The soft spot is the advertised interpretation. Theorems 4.4, 4.5, and 7.1 are moment-level CLTs for the empirical particle measure on one-dimensional slices. The step from those to 'the height function is the GFF plus an independent Brownian motion' is made in prose: Section 4.2 says 'one can verify' and invokes an unspecified diffeomorphism from the liquid region to the upper half-plane. That is not a proof of functional convergence of the two-dimensional height function, and the independence between the GFF and Brownian components is asserted, not derived. The stress-test note is right. This is not a manufactured concern: it is the difference between the abstract and the theorems.

It is also worth noting that the critical-variance results are largely Proposition 3.2 plus known theorems from [BG15] and [BG18]; the genuinely new work is the fixed-variance LLN/CLT. The limit-shape arctic curves in Section 5.4 depend on an unproved root-uniqueness condition, so the marked density plots are conditional on that.

For a referee: the core technical contribution deserves a serious round of review. The paper should not be desk-rejected, but the authors should be asked to either prove the Brownian identification or downgrade the advertised height-function statement to moment-level convergence, and to supply the omitted details in Theorem 4.5 and Section 6.2. Conditional acceptance is the right call.

Referee Report

4 major / 6 minor

Summary. This paper studies domino tilings of the Aztec diamond of size M with one-periodic i.i.d. random edge weights, encoded as random parameters β_i. The two regimes studied are: (i) critical vanishing variance, Var(β) ∼ σ²/M, where the paper advertises that unrescaled height fluctuations are the sum of the Gaussian Free Field and an independent one-dimensional Brownian motion; and (ii) fixed variance, where the paper advertises fluctuations of order √M described by Brownian motion alone. The technical core is a substantial extension of the Schur generating function method: an LLN (Theorem 5.6) based on the N-th root of the Schur generating function, a CLT (Theorem 6.2) at the fluctuation scale N^{k+1/2}, and applications to the Aztec diamond through the annealed Schur generating function computed in Proposition 3.2. The rigorous results are moment-level CLTs for the moments p_k of the particle measure along slices: Theorems 4.4, 4.5, and 7.1. Theorems 1.1 and 1.2 are informal versions that describe the full height function.

Significance. If the advertised field-level statements were fully proved, this would be a significant contribution. The annealed Schur generating function formula (3.1) is elegant and makes the random-environment model tractable, and the extension of the LLN/CLT framework from logarithms to N-th roots broadens the method in a way that is likely to be used elsewhere, including random matrix and free probability settings. The paper also contains explicit covariance formulas, free cumulant expressions, a random matrix degeneration, and concrete predictions for arctic curves; these are valuable even before the field-level interpretation is completed. The main weakness is that the headline statements about the full height function are stronger than the theorems that are actually proved, and the missing identification is explicitly acknowledged in the text rather than supplied.

major comments (4)

[4.2] The advertised decomposition of the height function fluctuations into an independent GFF plus one-dimensional Brownian motion is not proved. Theorems 4.4 and 4.5 are moment-level CLTs for the random variables p_k on individual slices, not functional convergence of the two-dimensional height function. The bridge in Section 4.2 is the sentence 'One can verify that this fluctuation term is interpreted as coming from ... one-dimensional Brownian motion.' No coordinate diffeomorphism from the liquid region to the upper half-plane is specified, no Brownian motion or time parameter is identified, and no proof is given that the two components are independent. Moment-level CLTs on a one-dimensional slice do not imply the field-level statement, so the central claim of Theorem 1.1 and of the abstract is currently unsupported. This is load-bearing and needs to be either proved or explicitly downgraded to a conjecture.
[6.4] The displayed computation after 'G^{(k,l)}(1_N)≈' contains an unreadable corrupted string (the repeated 'rro' symbols), so the proof of Theorem 6.2 cannot be checked. This is a principal contribution of the paper and is used directly in the proof of Theorem 7.1. Section 6.2 also states that the argument is only a sketch ('we do not repeat all the details ... but rather provide a sketch'). The corrupted formula and the sketch-level presentation must be repaired with a complete proof of Theorem 6.2.
[Theorem 4.5] The multilevel CLT is stated with covariance formula (4.5), but its proof is omitted: the text ends with 'We omit further details of the proof.' This is not a purely cosmetic omission. The multilevel covariance in (4.5) is the only object used in Section 4.2 to produce the Brownian-motion term and the claimed independence; without a proof of (4.5), even the moment-level input to the field-level decomposition is missing. Remark 7.2 similarly announces a multilevel analogue for the fixed-variance case without proof ('straightforwardly establish'). These gaps need to be filled or the claims explicitly marked as conjectural.
[5.4] The application of Proposition 5.12 to the Aztec diamond asserts that the limiting density is given by (5.20), where z solves equation (5.19), and Figure 7 presents arctic curves and densities based on this equation. No proof is given that (5.19) has a unique root in the upper half-plane for the random-environment F(z; alpha^{-1}-1), nor that the hypotheses of [BK18, Lemma 4.1 and Theorem 4.3] are satisfied for this non-explicit F. For non-discrete B the paper itself says the equation can only be solved numerically. Since the LLN/limit-shape claim in Theorem 1.2 depends on identifying the limiting measure from its moments, this is a further gap that should be addressed, either by proof or by an explicit conjecture.

minor comments (6)

[6] The heading 'Cental Limit Theorem' should be 'Central Limit Theorem'.
[4.1] In the paragraph before (4.1), 'convreges' should be 'converges'; also the compact-support assumption is stated in prose but is not included in the displayed assumptions (4.1)-(4.2), which can confuse the statement of Theorem 4.4.
[2.3] Proposition 2.5 contains 'this measuse admits' - 'measuse' should be 'measure'.
[5.4] In the proof of Proposition 5.12, 'This, only the expression ... contributes' should read 'Thus, only the expression ... contributes'.
[4.2] The displayed GFF covariance -1/(2 pi^2) log((z-w)/(zbar-w)) appears to have the wrong denominator; for the Dirichlet Green function in the upper half-plane one expects z - wbar. Please correct or clarify the convention.
[6.4] The last sentence of Section 6.4 says 'This completes the proof of Theorem 7.1', but the section is proving Theorem 6.2; correct the cross-reference.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the random-environment model is defined independently, the annealed Schur generating function is derived from the model, and the proved CLTs follow from that explicit generating function; the unproved Brownian interpretation is an evidence gap, not a circular reduction.

full rationale

The paper's derivation chain is self-contained in the relevant sense. The model is specified by i.i.d. random weights in Definition 3.1, independently of the asymptotic conclusions. Proposition 3.2 derives the annealed Schur generating function directly from the model via Proposition 2.5 and independence of the environment. The decreasing-variance results (Theorems 4.2, 4.4, 4.5) are obtained by combining this explicit generating function with the published parameter-free results of [BG15] and [BG18]; the limiting covariance depends on the input parameter sigma^2 = lim M Var(beta), which is an assumed model quantity rather than a fitted parameter. The fixed-variance results (Theorem 5.6, 6.2, 7.1) are proved in the paper from the stated asymptotics of the Schur generating function, with prior lemmas cited from published work. The advertised interpretation in Section 4.2 that the new covariance term 'is interpreted as coming from ... one-dimensional Brownian motion' is asserted rather than proved, and Remark 7.2 announces a multilevel analogue without proof; but this is an unproved strengthening or interpretive step, not a case where a prediction reduces to its input by construction. No equation is shown to equal its own input, and no fitted quantity is renamed as a prediction. The paper's heavy reliance on the authors' earlier works is legitimate because those works are published derivations with stated assumptions that do not include the present paper's conclusions. Therefore there is no circularity.

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

The central claims rest on standard Schur function machinery and on several explicitly stated but unproved assumptions: finite moments of the environment, root uniqueness for the limit-shape equation, and the Brownian-motion identification of the new covariance term. No free parameters are fitted to data, and no new physical entities are introduced.

assumptions (4)

domain assumption The distribution B (or B_M) has finite moments of all orders and, in the fixed-variance regime, is independent of M.
Definition 3.1 and Proposition 5.12 assume this to justify moment expansions and the N-th root limit.
ad hoc to paper Equation (5.19), F(z; alpha^{-1}-1)=alpha^{-1} y, has a unique root in the upper half-plane for each (alpha,y), or at least the limiting density formula (5.20) is valid.
Used in Section 5.4 to turn moment limits into explicit limit shapes and arctic curves (Figure 7); uniqueness is not proved for general distributions B.
ad hoc to paper The covariance term in Theorem 4.5 is a deterministic pushforward of one-dimensional Brownian motion after a suitable conformal map from the liquid region to the upper half-plane.
Section 4.2 says 'One can verify' and gives no proof; the Brownian motion interpretation in the abstract-level claims depends on this identification.
standard math The asymptotic theorems of [BG15] and [BG18], including the LLN and CLT for Schur generating functions used as black boxes, are correct.
The paper invokes these as established results, in particular Theorems 2.6 and 2.7 in Section 2.4.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Domino Tilings of the Aztec Diamond in Random Environment and Schur Generating Functions." pith.science (2026). https://pith.science/paper/RLU6L6KI

@misc{pith2026250708560,
  author       = {Pith},
  title        = {Pith review of: Domino Tilings of the Aztec Diamond in Random Environment and Schur Generating Functions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RLU6L6KI}},
  note         = {Machine review of arXiv:2507.08560}
}

abstract

We study the asymptotic behavior of random domino tilings of the Aztec diamond of size $M$ in a random environment, where the environment is a one-periodic sequence of i.i.d. random weights attached to domino positions (i.e., to the edges of the underlying portion of the square grid). We consider two cases: either the variance of the weights decreases at a critical scale $1/M$, or the distribution of the weights is fixed. In the former case, the unrescaled fluctuations of the domino height function are governed by the sum of a Gaussian Free Field and an independent Brownian motion. In the latter case, we establish fluctuations on the much larger scale $\sqrt M$, given by the Brownian motion alone. To access asymptotic fluctuations in random environment, we employ the method of Schur generating functions. Moreover, we substantially extend the known Law of Large Numbers and Central Limit Theorems for particle systems via Schur generating functions in order to apply them to our setting. These results might be of independent interest.

Figures

Figures reproduced from arXiv: 2507.08560 by the authors.

**Figure 1.** A uniformly random domino tiling of the Aztec diamond of size M = 200. In the present work we move in a different direction by allowing the weights themselves to be random. We focus on what is arguably the simplest nontrivial setting: a one-periodic model in which the parameters W1,W2, . . . ,WM are i.i.d.1 with a common distribution B (see Section 1.2 below for a precise definition, and [PITH_FULL_IMAGE:figures/… view at source ↗

**Figure 2.** Left: One-periodic weights Wi attached to the edges of the Aztec diamond graph of size M = 4. The edges which are not labeled have weight 1. We assume that Wi ’s form an i.i.d. random sequence with common distribution B. Right: An example of a dimer configuration on the Aztec diamond graph, and the corresponding domino tiling. Given the weights Wi , this configuration has conditional probability proportional to W3… view at source ↗

**Figure 3.** A domino tiling of the Aztec diamond of size M = 4 together with the corresponding height function and its 3D plot. Interactive 3D plots of domino height functions is available online at [Pet25a]. The height function changes by ±1 along the domino edges, and by ±3 across the dominos. The sign of the change depends on the parity with respect to the checkerboard coloring of the cells of Z2 . The height function (up to… view at source ↗

Figures from the paper (4 more)

**Figure 4.** Figure 4: Left: A random domino tiling of the Aztec diamond of size M = 200 with i.i.d. Bernoulli weights with P(Wi = 1 2 ) = P(Wi = 5) = 1 2 . Right: A random domino tiling of the Aztec diamond of size M = 200 with fixed nonrandom weights Wi = 1 2 and Wi = 5 for odd and even i,…

**Figure 5.** Figure 5: A random domino tiling of the Aztec diamond of size M = 300 with i.i.d. continuous uniform random weights Wi on [0, 2]. The right image is a grayscale version with eight shades depending on the parity of the dominos with respect to the lattice. See also [PITH_FULL_IMA…

**Figure 6.** Figure 6: Correspondence between domino tilings of the Aztec diamond and sequences of signatures (here, M = 4). We single out two types of dominos (red and green in the coloring in [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]

**Figure 7.** Figure 7: Arctic curves (top row) and plots of the limiting density (5.20) (bottom row) for domino tilings of the Aztec diamond with constant edge weights Wi ≡ 1 (left), i.i.d. Bernoulli edge weights with P(Wi = 1 2 ) = P(Wi = 5) = 1 2 (middle), and i.i.d. uniform edge weights …

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

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

Dimers with layered disorder
math.PR 2025-07 conditional novelty 8.0 of 10

Layered disorder creates an essential singularity in the dimer free energy and modifies the liquid-gas critical exponent continuously.
Quenched and Annealed CLTs for the one-periodic Aztec diamond in random environment
math.PR 2025-10 conditional novelty 7.0 of 10

Quenched height fluctuations in random-environment one-periodic Aztec diamonds converge almost surely to the Gaussian Free Field; annealed fluctuations are Gaussian with environment-dependent covariances.

Reference graph

Works this paper leans on

53 extracted references · 43 canonical work pages · cited by 2 Pith papers

[1]

Berestycki, B

N. Berestycki, B. Laslier, and G. Ray. Dimers and Imaginary geometry . Ann. Probab. , 48(1):1--52, 2020

work page 2020
[2]

Domino tilings of the Aztec diamond with doubly periodic weightings

T. Berggren. Domino tilings of the Aztec diamond with doubly periodic weightings . Ann. Probab. , 49(4):1965--2011, 2021. arXiv:1911.01250 [math.PR]

work page Pith review arXiv 1965
[3]

Geometry of the doubly periodic Aztec dimer model

T. Berggren and A. Borodin. Geometry of the doubly periodic Aztec dimer model . arXiv preprint , 2023. arXiv:2306.07482 [math.PR]

work page Pith review arXiv 2023
[4]

Berggren and A

T. Berggren and A. Borodin. Crystallization of the Aztec diamond . arXiv preprint , 2024. arXiv:2410.04187 [math-ph]

arXiv 2024
[5]

Berggren and M

T. Berggren and M. Nicoletti. Gaussian Free Field and Discrete Gaussians in Periodic Dimer Models . arXiv preprint , 2025. arXiv:2502.07241 [math.PR]

arXiv 2025
[6]

Perfect t-embeddings of uniformly weighted Aztec diamonds and tower graphs

T. Berggren, M. Nicoletti, and M. Russkikh. Perfect t-embeddings of uniformly weighted Aztec diamonds and tower graphs . Int. Math. Res. Not. , 2024(7):5963--6007, 2024. arXiv:2303.10045 [math-ph]

work page Pith review arXiv 2024
[7]

Betea, C

D. Betea, C. Boutillier, J. Bouttier, G. Chapuy, S. Corteel, and M. Vuleti \'c . Perfect sampling algorithm for Schur processes . Markov Processes and Related Fields , 24(3):381--418, 2017. arXiv:1407.3764 [math.PR]

arXiv 2017
[8]

A. I. Bobenko and N. Bobenko. Dimers and M-Curves: Limit Shapes from Riemann Surfaces . arXiv preprint , 2024. arXiv:2407.19462 [math-ph]

arXiv 2024

Show all 53 references

[9]

A. Borodin. Schur dynamics of the Schur processes . Adv. Math. , 228(4):2268--2291, 2011. arXiv:1001.3442 [math.CO]

2011 arXiv
[10]

Borodin and P

A. Borodin and P. Ferrari. Anisotropic growth of random surfaces in 2+1 dimensions . Commun. Math. Phys. , 325:603--684, 2014. arXiv:0804.3035 [math-ph]

2014 arXiv
[11]

Borodin, V

A. Borodin, V. Gorin, and E. Strahov. Product matrix processes as limits of random plane partitions . Int. Math. Res. Not. , 2020(20):6713--6768, 2020. arXiv:1806.10855 [math-ph]

2020 arXiv
[12]

Boutillier and B

C. Boutillier and B. de Tilière. Fock's dimer model on the Aztec diamond . arXiv preprint , 2024. arXiv:2405.20284 [math.PR]

2024 arXiv
[13]

Bufetov and V

A. Bufetov and V. Gorin. Representations of classical Lie groups and quantized free convolution . Geometric And Functional Analysis , 25(3):763--814, 2015. arXiv:1311.5780 [math.RT]

2015 arXiv
[14]

Bufetov and V

A. Bufetov and V. Gorin. Fluctuations of particle systems determined by Schur generating functions . Adv. Math. , 338:702--781, 2018. arXiv:1604.01110 [math.PR]

2018 arXiv
[15]

Bufetov and V

A. Bufetov and V. Gorin. Fourier transform on high-dimensional unitary groups with applications to random tilings . Duke Math. J. , 168(13):2559--2649, 2019. arXiv:1712.09925 [math.PR]

2019 arXiv
[16]

Bufetov and A

A. Bufetov and A. Knizel. Asymptotics of random domino tilings of rectangular Aztec diamonds . Ann. Inst. H. Poincare Probab. Statist. , 54(3):1250--1290, 2018. arXiv:1604.01491 [math.PR]

2018 arXiv
[17]

Bufetov and P

A. Bufetov and P. Zografos. Asymptotics of Harish-Chandra transform and infinitesimal freeness . arXiv preprint , 2024. arXiv:2412.09290 [math.PR]

2024 arXiv
[18]

Chelkak and S

D. Chelkak and S. Ramassamy. Fluctuations in the Aztec diamonds via a space-like maximal surface in Minkowski 3-space . Confl. Math. , 16:1--17, 2024. arXiv:2002.07540 [math-ph]

2024 arXiv
[19]

Chhita and K

S. Chhita and K. Johansson. Domino statistics of the two-periodic Aztec diamond . Adv. Math. , 294:37--149, 2016. arXiv:1410.2385 [math.PR]

2016 arXiv
[20]

Chhita, K

S. Chhita, K. Johansson, and B. Young. Asymptotic domino statistics in the Aztec diamond . Ann. Appl. Probab. , 25(3):1232--1278, 2015. arXiv:1212.5414 [math.PR]

2015 arXiv
[21]

Chhita and B

S. Chhita and B. Young. Coupling functions for domino tilings of Aztec diamonds . Adv. Math. , 259:173--251, 2014. arXiv:1302.0615 [math.CO]

2014 arXiv
[22]

H. Cohn, R. Kenyon, and J. Propp. A variational principle for domino tilings. Jour. AMS , 14(2):297--346, 2001. arXiv:math/0008220 [math.CO]

2001 arXiv
[23]

Collins, J

B. Collins, J. Novak, and P. \'S niady. Semiclassical asymptotics of GL_N( C ) tensor products and quantum random matrices . Sel. Math. New Ser. , 24(3):2571--2623, 2018. arXiv:1611.01892 [math.RT]

2018 arXiv
[24]

Cuenca and M

C. Cuenca and M. Do ega. Discrete N-particle systems at high temperature through Jack generating functions . arXiv preprint , 2025. arXiv:2502.13098 [math.PR]

2025
[25]

Di Francesco and M.F

P. Di Francesco and M.F. Lapa. Arctic curves in path models from the tangent method . J. Phys. A: Math. Theor. , 51(15):155202, 2018. arXiv:1711.03182 [math-ph]

2018 arXiv
[26]

Di Francesco and R

P. Di Francesco and R. Soto-Garrido. Arctic curves of the octahedron equation . J. Phys. A , 47(28):285204, 2014. arXiv:1402.4493 [math-ph]

2014 arXiv
[27]

Duits and A

M. Duits and A. Kuijlaars. The two periodic Aztec diamond and matrix valued orthogonal polynomials . Journal of the European Mathematical Society , 23(4):1075--1131, 2020. arXiv:1712.05636 [math.PR]

2020 arXiv
[28]

Duits and R

M. Duits and R. Van Peski. Gamma-disordered dimers and integrable polymers , 2025. in preparation

2025
[29]

Elkies, G

N. Elkies, G. Kuperberg, M. Larsen, and J. Propp. Alternating-sign matrices and domino tilings. Jour. Alg. Comb. , 1(2-3):111--132 and 219--234, 1992

1992
[30]

V. Gorin. Lectures on random lozenge tilings. Cambridge Studies in Advanced Mathematics. Cambridge University Press , 2021

2021
[31]

Gorin and Y

V. Gorin and Y. Sun. Gaussian fluctuations for products of random matrices . Am. J. Math. , 144(2):287--393, 2022. arXiv:1812.06532 [math.PR]

2022 arXiv
[32]

Guionnet and M

A. Guionnet and M. Ma \" da. A Fourier view on the R-transform and related asymptotics of spherical integrals . J. Funct. Anal. , 222(2):435--490, 2005. arXiv:math/0406121 [math.PR]

2005 arXiv
[33]

J. Huang. Law of Large Numbers and Central Limit Theorems through Jack Generating Functions . Adv. Math. , 380:107545, 2021

2021
[34]

Jockusch, J

W. Jockusch, J. Propp, and P. Shor. Random Domino Tilings and the Arctic Circle Theorem . arXiv preprint , 1998. arXiv:math/9801068 [math.CO]

1998 arXiv
[35]

Johansson

K. Johansson. Non-intersecting paths, random tilings and random matrices . Probab. Theory Relat. Fields , 123(2):225--280, 2002. arXiv:math/0011250 [math.PR]

2002 arXiv
[36]

Johansson

K. Johansson. The arctic circle boundary and the A iry process. Ann. Probab. , 33(1):1--30, 2005. arXiv:math/0306216 [math.PR]

2005 arXiv
[37]

Johansson and E

K. Johansson and E. Nordenstam. Eigenvalues of GUE minors . Electron. J. Probab. , 11(50):1342--1371, 2006. arXiv:math/0606760 [math.PR]; Erratum: Electron. J. Probab. 12 (2007), no. 37, 1048--1051

2007 arXiv
[38]

R. Kenyon. Dominos and the Gaussian Free Field . Ann. Probab. , 29(3):1128--1137, 2001. arXiv:math-ph/0002027

2001 arXiv
[39]

R. Kenyon. Height fluctuations in the honeycomb dimer model. Commun. Math. Phys. , 281(3):675--709, 2008. arXiv:math-ph/0405052

2008 arXiv
[40]

Kenyon and A

R. Kenyon and A. Okounkov. Limit shapes and the complex B urgers equation. Acta Math. , 199(2):263--302, 2007. arXiv:math-ph/0507007

2007 arXiv
[41]

Kenyon, A

R. Kenyon, A. Okounkov, and S. Sheffield. Dimers and amoebae. Ann. Math. , 163:1019--1056, 2006. arXiv:math-ph/0311005

2006 arXiv
[42]

Kenyon and I

R. Kenyon and I. Prause. Limit shapes from harmonicity: dominos and the five vertex model . J. Phys. A: Math. Theor. , 57(3):035001, 2024. arXiv:2310.06429 [math-ph]

2024 arXiv
[43]

S. Mason. Two-periodic weighted dominos and the sine-Gordon field at the free fermion point: I . arXiv preprint , 2022. arXiv:2209.11111 [math-ph]

2022 arXiv
[44]

Matsumoto and J

S. Matsumoto and J. Novak. A moment method for invariant ensembles . Electron. Res. Announc. Math. Sci. , 25:60--71, 2018

2018
[45]

Okounkov

A. Okounkov. Infinite wedge and random partitions . Selecta Math. , 7(1):57--81, 2001. arXiv:math/9907127 [math.RT]

2001 arXiv
[46]

Okounkov and N

A. Okounkov and N. Reshetikhin. Correlation function of Schur process with application to local geometry of a random 3-dimensional Young diagram . Jour. AMS , 16(3):581--603, 2003. arXiv:math/0107056 [math.CO]

2003 arXiv
[47]

Perret, Z

A. Perret, Z. Ristivojevic, P. Le Doussal, G. Schehr, and K. J. Wiese. Super-Rough Glassy Phase of the Random Field XY Model in Two Dimensions . Phys. Rev. Lett. , 109:157205, 2012. arXiv:1204.5685 [cond-mat.dis-nn]

2012 arXiv
[48]

L. Petrov. Asymptotics of Uniformly Random Lozenge Tilings of Polygons. Gaussian Free Field . Ann. Probab. , 43(1):1--43, 2015. arXiv:1206.5123 [math.PR]

2015 arXiv
[49]

L. Petrov. Domino tilings. Interactive 2D and 3D simulations . https://lpetrov.cc/domino/, 2025

2025
[50]

L. Petrov. Domino tilings of the Aztec diamond with random one-periodic edge weights. Interactive simulation . https://lpetrov.cc/simulations/2025-06-25-random-edges/, 2025

2025
[51]

Generalized domino-shuffling

James Propp. Generalized domino-shuffling. Theoretical Computer Science , 303(2-3):267--301, 2003. arXiv:math/0111034 [math.CO]

2003 arXiv
[52]

Speicher

R. Speicher. Free Probability Theory . In G. Akemann, J. Baik, and P. Di Francesco, editors, The Oxford Handbook of Random Matrix Theory , pages 452--470. Oxford University Press, 2011. arXiv:0911.0087 [math.PR]

2011 arXiv
[53]

W. P. Thurston. Conway's tiling groups . Amer. Math. Monthly , 97:757--773, 1990

1990

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [1]

Berestycki, B

N. Berestycki, B. Laslier, and G. Ray. Dimers and Imaginary geometry . Ann. Probab. , 48(1):1--52, 2020

work page 2020

[2] [2]

Domino tilings of the Aztec diamond with doubly periodic weightings

T. Berggren. Domino tilings of the Aztec diamond with doubly periodic weightings . Ann. Probab. , 49(4):1965--2011, 2021. arXiv:1911.01250 [math.PR]

work page Pith review arXiv 1965

[3] [3]

Geometry of the doubly periodic Aztec dimer model

T. Berggren and A. Borodin. Geometry of the doubly periodic Aztec dimer model . arXiv preprint , 2023. arXiv:2306.07482 [math.PR]

work page Pith review arXiv 2023

[4] [4]

Berggren and A

T. Berggren and A. Borodin. Crystallization of the Aztec diamond . arXiv preprint , 2024. arXiv:2410.04187 [math-ph]

arXiv 2024

[5] [5]

Berggren and M

T. Berggren and M. Nicoletti. Gaussian Free Field and Discrete Gaussians in Periodic Dimer Models . arXiv preprint , 2025. arXiv:2502.07241 [math.PR]

arXiv 2025

[6] [6]

Perfect t-embeddings of uniformly weighted Aztec diamonds and tower graphs

T. Berggren, M. Nicoletti, and M. Russkikh. Perfect t-embeddings of uniformly weighted Aztec diamonds and tower graphs . Int. Math. Res. Not. , 2024(7):5963--6007, 2024. arXiv:2303.10045 [math-ph]

work page Pith review arXiv 2024

[7] [7]

Betea, C

D. Betea, C. Boutillier, J. Bouttier, G. Chapuy, S. Corteel, and M. Vuleti \'c . Perfect sampling algorithm for Schur processes . Markov Processes and Related Fields , 24(3):381--418, 2017. arXiv:1407.3764 [math.PR]

arXiv 2017

[8] [8]

A. I. Bobenko and N. Bobenko. Dimers and M-Curves: Limit Shapes from Riemann Surfaces . arXiv preprint , 2024. arXiv:2407.19462 [math-ph]

arXiv 2024