pith. sign in

arxiv: 2211.14365 · v3 · submitted 2022-11-25 · 🧮 math.PR · math-ph· math.MP

A dichotomy theory for the height functions of the BKT transition

Piet Lammers This is my paper

Pith reviewed 2026-05-24 11:04 UTC · model grok-4.3

classification 🧮 math.PR math-phmath.MP
keywords BKT transitionheight functionslocalizationdelocalizationtwo-point functionvariance growtheffective temperaturephase diagram
0
0 comments X

The pith

Localized BKT height functions have exponentially decaying covariances while delocalized ones grow in variance at least as c log n with universal c>0.

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

The paper establishes a dichotomy for discrete height functions tied to the Berezinskii-Kosterlitz-Thouless transition at slope zero. When the model is localised the covariance between any two points decays exponentially with distance. When delocalised the variance of the height at distance n from the boundary is bounded below by c log n, where c is a positive constant independent of temperature. This forces an effective-temperature gap: values strictly between 0 and c are impossible. The delocalised regime is closed in the phase diagram and therefore contains the transition point itself.

Core claim

If the model is localised then the two-point function decays exponentially fast in the distance between the points; if delocalised then the variance grows at least as c log n with universal c>0 independent of temperature, implying that effective temperature must jump from 0 to at least c at the transition; the delocalised phase is closed and includes the transition point.

What carries the argument

The localisation-delocalisation dichotomy for the height functions, with the universal lower bound on logarithmic variance growth serving as the mechanism that enforces the effective-temperature gap.

If this is right

  • Effective temperatures strictly between 0 and c are forbidden at the transition.
  • The transition point belongs to the delocalised phase.
  • The localisation-delocalisation transition is equivalent to the BKT transition in the dual XY and Villain models.
  • Values of the effective temperature in (0, c) cannot occur anywhere in the phase diagram under the given topology.

Where Pith is reading between the lines

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

  • The same variance-gap argument may apply to other two-dimensional models whose height representations exhibit a BKT-type transition.
  • The universal constant c could be computed explicitly once a specific instance of the height function is fixed.
  • The dichotomy supplies a new criterion for identifying the location of the transition without direct computation of the free energy.

Load-bearing premise

The definitions of the localised and delocalised phases together with the topology on the phase diagram are such that the variance lower bound directly produces a jump in effective temperature.

What would settle it

An explicit height-function model at the transition point whose variance grows as k log n with 0 < k < c, or whose effective temperature takes a value inside (0, c), would falsify the claimed gap.

Figures

Figures reproduced from arXiv: 2211.14365 by Piet Lammers.

Figure 1
Figure 1. Figure 1: Left: A sample at low temperature from the discrete Gaussian model. The sample looks flat with a few local excitations. Right: A sample at high temperature from the same model. The surface looks rougher and the heights tend to move away from zero. years. The purpose of this article is to investigate the nature of the phase transition which occurs for height functions. There are two phases: either the varia… view at source ↗
Figure 2
Figure 2. Figure 2: The (simplified version of the) observable pn(V ) is defined as the probability of seeing a circuit at height ≥ 1 in the small annulus, in the measure with height 0 imposed on the boundary of the large box. The small annulus and the large box both scale linearly with n. Thus, the observable measures the ability of the model to transition from one height to another at the macroscopic scale. The formal defin… view at source ↗
Figure 3
Figure 3. Figure 3: Schematic rendering of the main results in the phase diagram, realised as the topological space (Φ, T ). The infinite-dimensional space is projected onto paper in such a way that the x-coordinate coincides precisely with T(V ) := (V (1) − V (0))−1 . Unlike perhaps suggested, it is not proved that Loc[Φ] and Deloc[Φ] are connected in this topology. Some existing localisation-delocalisation results are drawn… view at source ↗
Figure 4
Figure 4. Figure 4: The coarse-graining inequalities separate the four phases of the random-cluster model through three dichotomies. The first coarse-graining inequality has a primal and a dual version. Since the subcritical and the supercritical regime do not occur for height functions, it is the second coarse￾graining inequality that describes the localisation-delocalisation transition for height functions. full knowledge o… view at source ↗
Figure 5
Figure 5. Figure 5: A sample from µΛ,τ,ξ with the percolations, relative to the reference height a = 0. Only the signs of the heights are given. The heights on Λ are represented by circles; the heights on the boundary by squares. Left: The primal percolation (black) is L  0 , the dual percolation (aquamarine) is L0. Observe that each edge in L  0 connects vertices of the same sign. The percolation L  ≤0 (not drawn) contain… view at source ↗
Figure 6
Figure 6. Figure 6: Quads with boundary conditions as appearing in Lemma 6.3. The boundary is thick where ξ equals zero and thin where ξ equals one. Left: A wide quad with height-one boundary conditions on the left and right. The paths to infinity for the technical condition are also drawn. Middle: The union of the boundary condition with the symmetrised version of the geometric domain. The technical condition ensures that al… view at source ↗
Figure 7
Figure 7. Figure 7: A T-quad Q: the thin line is TopQ, the rest of ∂Q is thick. To conclude, we show that in fact at least one of the two events must almost surely occur for any quad with the given boundary height function ˆξ. Suppose that instead both events do not occur. Focus on the left event. Since this event does not occur, there must exist a primal path p 0 which is open for L  0 with the following properties: • The f… view at source ↗
Figure 8
Figure 8. Figure 8: The generic symmetrisation argument. For F, we chose the vertical crossing event of the rectangle on the left. The point z = (x, y) marks the symmetry point; ΣF is the crossing event of the rectangle on the right. The boundary conditions (Λ, τ, ξ) and (Λ0 , τ 0 , ξ00) are represented by the figures on the left and right respectively. The change in boundary conditions increases K0 and decreases K? 0 . Recal… view at source ↗
Figure 9
Figure 9. Figure 9: A T-quad with the universes Uout, Umid, and Uin. see [PITH_FULL_IMAGE:figures/full_fig_p042_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Arm∗ [x, y, m], Bridge∗ [x, y, m], and Quasi∗ [x, y, m, m0 ] 1. ω 0 ∈ Arm∗ [x, y, m0 ] and ω 0 contains an edge traversing LeftBox3β(x,y,m) , 2. ω 00 ∈ Arm∗ [x, y, m0 ] and ω 00 contains an edge traversing RightBox3β(x,y,m) . Our objective is to derive Lemma 7.3, which asserts that B percolates on the macroscopic scale. Suppose that we want to derive Perc[m, α, U] for m ≈ N. To prove this, we must essenti… view at source ↗
Figure 11
Figure 11. Figure 11: Top left: The horizontal crossing has probability at least α by assumption. Middle left: We claim that µ(H(A)) ≥ α 2 c 2 meso/2. Bottom left: The event H(A) guarantees a vertical crossing of the middle rectangle by B. Top right: Intersections of events of the form H(A) guarantee vertical crossings of taller rectangles by B. Bottom right: Horizontal crossings of wider rectangles are created by combining ev… view at source ↗
Figure 12
Figure 12. Figure 12: The paths L ± trace (part of) the boundary of V ±. The quad Q is defined such that TopQ ⊂ L+ and BottomQ ⊂ L−, and such that LeftQ and RightQ, which are both contained in L≤1, run along the boundary of VL and VR respectively. The event Z guarantees that LeftQ and RightQ remain strictly on the left and right respectively of the middle square, and also makes the technical condition work. The event H(A) occu… view at source ↗
Figure 13
Figure 13. Figure 13: The event DoubleArm[R, m]. For the definition it does not matter whether or not the two boxes overlap. Corollary 7.10. Let U and U 0 denote two universes with U 0 strictly contained in U. Then for any 1 ≤ m ≤ N/90 and for any 1 ≤ n < n0 ≤ 16N, we have ∀R ∈ Rect[n × m, U], µ(B ∈ Ver∗ {R}) ≥ δ ∀R ∈ Rect[n 0 × m, U], µ(B ∈ Hor∗ {R}) ≥ δ 0 ) =⇒ Perc[9m, α, U0 ]; α :=  (δδ0 ) 2d 10m n0−n e c 2 meso/2 18 . 7.… view at source ↗
Figure 14
Figure 14. Figure 14: The event B ∈ Quasi∗ [x, y, mk, mk+1] contains the intersection of four other events due to the corridor construction [KT23, Lemma 2]. If a primal edge is open for K? 0 , then its dual edge and the six surrounding dual edges are open for B; see Proposition 2.18. denote the unique rectangles such that the lower box Boxβ(x, yb, mk+1) and the upper box Boxβ(x, yt , mk+1) respectively in the definition of Dou… view at source ↗
Figure 15
Figure 15. Figure 15: Two paths γ, γ˜ ∈ Γ. Recall the square root trick from Section 4: if P is any percolation measure satisfying the FKG inequality and if A1, . . . , An are increasing events with ∪kAk =: A, then max{P(A1), . . . , P(An)} ≥ fn(P(A)); fn(x) := 1 − √n 1 − x. Note that fa ◦ fb = fab and that f8·8·9(1 − δaspect) > 1 2 because δaspect := 2−600 . We first introduce some notation which will make it easier to descri… view at source ↗
Figure 16
Figure 16. Figure 16: Crossing events and deviation events. The continuous and dashed lines represent a path in the + and the − set respectively on either side of the figure. The symmetry point is also marked. Observe that on either side any + path must intersect any − path. Proof. We shall prove the claim for the case that η represents γ; the other cases are the same. Let Σ ∈ Σ  denote the symmetry which rotates the plane ar… view at source ↗
Figure 17
Figure 17. Figure 17: A path γ ∈ Γ 00. The sub-paths γ ± have been drawn as a continuous line. The definition of Γ 00 implies that the paths γ ± remain in their respective rectangles and that they enter and exit the rectangles through the marked targets. 8. The second coarse-graining inequality This section formally introduces the annulus observable and proves the second coarse￾graining inequality (Lemma 1). The observable is … view at source ↗
Figure 18
Figure 18. Figure 18: The observable pn measures the probability of seeing an L1- circuit through the smaller annulus, given zero boundary conditions on a circuit through the larger annulus. The observable is defined as the supremum of this probability over all such boundary circuits; this is necessary for technical reasons. are defined at each scale n ∈ Z≥1 satisfy, for each n ∈ Z≥1000, the equation p20kn(V ) ≤ (pn(V )/cdicho… view at source ↗
Figure 19
Figure 19. Figure 19: The proof of (34). Thin lines indicate L≤1; thick lines L≤0. Define the event H(L0) :=  L0 ∈ Hor∗ {[[40kn]] × [−3n, −2n]} ∩ Hor∗ {[[40kn]] × [2n, 3n]} [PITH_FULL_IMAGE:figures/full_fig_p058_19.png] view at source ↗
Figure 20
Figure 20. Figure 20: The proof of (35). Thin lines indicate L1; thick lines L0. Focus on (33). Lemma 5.2 implies µΛ,τ,0(H3n(L≤1)|A∗ k,n(L1, n)) ≥ c 320k via by exploring the annulus circuits from the inside, then applying the lemma with the ideas in Remark 5.3 and monotonicity in domains (observing that this time Λ ⊂ Λ160kn). Now let c denote the constant from Corollary 7.2. That corollary implies that µΛ,τ,0(H5n(L≤0)|H3n(L≤1… view at source ↗
read the original abstract

This text considers the discrete height functions associated with the Berezinskii--Kosterlitz--Thouless transition (BKT) at slope zero. Our main results are as follows. * Sharpness: If the model is localised, then the two-point function (covariance) decays exponentially fast in the distance between the points. * Effective temperature gap: If the model is delocalised, then the variance grows at least as $c\log n$, where $n$ is the distance to the boundary and $c>0$ a universal constant not depending on the temperature. Thus, the effective temperature must jump from $0$ to at least $c$ at the transition point; values in the interval $(0,c)$ are forbidden. * Delocalisation at the transition point: The delocalised phase includes the transition point, in the sense that it is a closed set in the phase diagram in the appropriate topology. These results contribute to the understanding of the regime at and around the transition point which remained largely unexplored. In a follow-up paper, the sharpness derived here is used to establish that the localisation-delocalisation transition is equivalent to the BKT transition in the dual XY and Villain models.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 1 minor

Summary. The paper develops a dichotomy theory for discrete height functions of the Berezinskii-Kosterlitz-Thouless (BKT) transition at slope zero. It claims three main results: (i) localization implies exponential decay of the two-point covariance function; (ii) delocalization implies that the variance grows at least as c log n (n distance to boundary) for a universal c > 0 independent of temperature, thereby forcing an effective-temperature gap from 0 to at least c at the transition; (iii) the delocalized phase is closed in the appropriate topology on the phase diagram and therefore contains the transition point. These statements are positioned as enabling a follow-up equivalence between the localization-delocalization transition and the BKT transition in dual XY/Villain models.

Significance. If the central claims hold with the stated universality and without circularity in the phase definitions, the work would supply a sharp, parameter-free characterization of the BKT regime for height functions that has remained largely open. The universal lower bound on variance growth and the closure property would constitute a genuine advance, directly supporting the claimed equivalence in the companion paper.

major comments (2)
  1. [Abstract / phase-diagram definition] Abstract (effective-temperature-gap statement): the implication that delocalization forces effective temperature to jump from 0 to at least c rests on the chosen definitions of 'localised'/'delocalised' and the topology on the phase diagram. The manuscript must explicitly verify (in the section introducing the phase diagram and effective temperature) that these notions are independent of the variance lower bound itself; otherwise the gap result is tautological rather than a theorem.
  2. [Abstract / topology on phase diagram] Abstract (delocalisation-at-transition statement): the claim that the delocalised phase is closed and contains the transition point requires that the topology be the natural one induced by the height-function model. The manuscript should supply a precise statement of this topology (likely in the section defining the phase diagram) and confirm that the closure property follows from the variance lower bound without additional assumptions.
minor comments (1)
  1. [Abstract] The abstract states the three results at a high level but supplies no proof sketches or references to the sections containing the arguments; adding one-sentence pointers to the relevant theorems would improve readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive report. The two major comments raise important points about potential circularity in the phase definitions and the need for an explicit topology. We address each below and will revise the manuscript accordingly to strengthen the presentation.

read point-by-point responses

  1. Referee: [Abstract / phase-diagram definition] Abstract (effective-temperature-gap statement): the implication that delocalization forces effective temperature to jump from 0 to at least c rests on the chosen definitions of 'localised'/'delocalised' and the topology on the phase diagram. The manuscript must explicitly verify (in the section introducing the phase diagram and effective temperature) that these notions are independent of the variance lower bound itself; otherwise the gap result is tautological rather than a theorem.

    Authors: The definitions are independent. Localization and delocalization are introduced in Section 2 via the qualitative behavior of the height function (whether the measure concentrates on bounded heights or not), prior to any variance statements. Effective temperature is defined separately as the lim sup of Var(h(x))/log dist(x,∂Λ). The lower bound theorem then shows that any delocalized regime forces this lim sup to be at least a universal c>0. We will add an explicit verification paragraph in the phase-diagram section of the revision confirming that none of these definitions presuppose the value of the lim sup, so the gap is a genuine theorem. revision: yes

  2. Referee: [Abstract / topology on phase diagram] Abstract (delocalisation-at-transition statement): the claim that the delocalised phase is closed and contains the transition point requires that the topology be the natural one induced by the height-function model. The manuscript should supply a precise statement of this topology (likely in the section defining the phase diagram) and confirm that the closure property follows from the variance lower bound without additional assumptions.

    Authors: We will supply the missing precise statement. The topology on the phase diagram is the one induced by weak convergence of the finite-volume Gibbs measures (equivalently, by convergence of all finite-dimensional marginals or of the effective temperatures). Under this topology the variance functional is lower semi-continuous. Consequently, if a sequence of delocalized parameters converges to a limit, the uniform lower bound c log n on variance passes to the limit, showing the limit point is delocalized. This argument uses only the variance theorem already proved and requires no extra assumptions. The revised manuscript will contain a formal subsection stating the topology and proving closure from the variance bound. revision: yes

Circularity Check

0 steps flagged

No circularity; results rest on independent mathematical arguments

full rationale

The paper states three main theorems (sharpness of localization via exponential decay, effective-temperature gap via variance lower bound, and closure of the delocalized phase) as derived results for discrete height functions. No quoted steps reduce by definition, by fitting a parameter then relabeling it a prediction, or by load-bearing self-citation whose content is unverified. The follow-up paper is cited only for a subsequent application, not to justify the present claims. The derivation chain is therefore self-contained against external benchmarks and receives the default non-circularity finding.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review yields no explicit free parameters, axioms, or invented entities; the results rest on standard definitions of height functions and BKT models from prior literature.

pith-pipeline@v0.9.0 · 5746 in / 1196 out tokens · 34455 ms · 2026-05-24T11:04:10.912469+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Forward citations

Cited by 1 Pith paper

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

  1. The impact of disorder and non-convex interactions on delocalisation of height functions

    math.PR 2026-04 unverdicted novelty 7.0

    Phase transitions in XY/Villain models and dual height functions persist under quenched disorder, and rough phases exist for annealed non-convex potentials like |∇h|^p with p≤2.

Reference graph

Works this paper leans on

43 extracted references · 43 canonical work pages · cited by 1 Pith paper · 2 internal anchors

  1. [1]

    1, 304--331

    Michael Aizenman, Matan Harel, and Ron Peled, Exponential decay of correlations in the 2D random field I sing model , Journal of Statistical Physics 180 (2020), no. 1, 304--331

    work page 2020

  2. [2]

    Michael Aizenman, Matan Harel, Ron Peled, and Jacob Shapiro, Depinning in integer-restricted G aussian fields and BKT phases of two-component spin models , arXiv preprint arXiv:2110.09498 (2021)

    work page arXiv 2021

  3. [3]

    C lassical systems , Soviet physics JETP 32 (1971), no

    VL Berezinskii, Destruction of long-range order in one-dimensional and two-dimensional systems having a continuous symmetry group I . C lassical systems , Soviet physics JETP 32 (1971), no. 3, 493--500

    work page 1971

  4. [4]

    Lebowitz, Surface tension, percolation, and roughening, Journal of Statistical Physics 29 (1982), no

    Jean Bricmont, Jean Raymond Fontaine, and Joel L. Lebowitz, Surface tension, percolation, and roughening, Journal of Statistical Physics 29 (1982), no. 2, 193--203

    work page 1982

  5. [5]

    a ggstr \

    Itai Benjamini, Olle H \"a ggstr \"o m, and Elchanan Mossel, On random graph homomorphisms into Z , Journal of Combinatorial Theory, Series B 78 (2000), no. 1, 86--114

    work page 2000

  6. [6]

    R enormalisation group flow at high temperature , arXiv preprint arXiv:2202.02286 (2022)

    Roland Bauerschmidt, Jiwoon Park, and Pierre-Fran c ois Rodriguez, The discrete G aussian model, I . R enormalisation group flow at high temperature , arXiv preprint arXiv:2202.02286 (2022)

    work page arXiv 2022

  7. [7]

    I nfinite-volume scaling limit at high temperature , arXiv preprint arXiv:2202.02287 (2022)

    , The discrete G aussian model, II . I nfinite-volume scaling limit at high temperature , arXiv preprint arXiv:2202.02287 (2022)

    work page arXiv 2022

  8. [8]

    3, 425--440

    R Brandenberger and CE Wayne, Decay of correlations in surface models, Journal of Statistical Physics 27 (1982), no. 3, 425--440

    work page 1982

  9. [9]

    2, 621--647

    Nishant Chandgotia, Ron Peled, Scott Sheffield, and Martin Tassy, Delocalization of uniform graph homomorphisms from Z^2 to Z , Communications in Mathematical Physics 387 (2021), no. 2, 621--647

    work page 2021

  10. [10]

    Hugo Duminil-Copin , Maxime Gagnebin, Matan Harel, Ioan Manolescu, and Vincent Tassion, The B ethe ansatz for the six-vertex and XXZ models: A n exposition , Probability Surveys 15 (2018), 102--130

    work page 2018

  11. [11]

    54, 2021, pp

    , Discontinuity of the phase transition for the planar random-cluster and P otts models with q> 4 , Annales Scientifiques de l'Ecole Normale Sup \'e rieure, vol. 54, 2021, pp. 1363--1413

    work page 2021

  12. [12]

    1, 315--347

    Hugo Duminil-Copin , Alexander Glazman, Ron Peled, and Yinon Spinka, Macroscopic loops in the loop O (n) model at N ienhuis' critical point , Journal of the European Mathematical Society 23 (2020), no. 1, 315--347

    work page 2020

  13. [13]

    Hugo Duminil-Copin , Matan Harel, Benoit Laslier, Aran Raoufi, and Gourab Ray, Logarithmic variance for the height function of square-ice, Communications in Mathematical Physics (2022), 1--36

    work page 2022

  14. [14]

    Paul Dario, Matan Harel, and Ron Peled, Random-field random surfaces, Probability Theory and Related Fields (2023), 1--68

    work page 2023

  15. [15]

    Hugo Duminil-Copin , Alex Karrila, Ioan Manolescu, and Mendes Oulamara, Delocalization of the height function of the six-vertex model, arXiv preprint arXiv:2012.13750 (2020)

    work page arXiv 2012

  16. [16]

    1, 75--99

    Hugo Duminil-Copin , Aran Raoufi, and Vincent Tassion, Sharp phase transition for the random-cluster and P otts models via decision trees , Annals of Mathematics 189 (2019), no. 1, 75--99

    work page 2019

  17. [17]

    1, 47--107

    Hugo Duminil-Copin , Vladas Sidoravicius, and Vincent Tassion, Continuity of the phase transition for planar random-cluster and P otts models with 1 q 4 , Communications in Mathematical Physics 349 (2017), no. 1, 47--107

    work page 2017

  18. [18]

    Hugo Duminil-Copin and Vincent Tassion, Renormalization of crossing probabilities in the planar random-cluster model, arXiv preprint arXiv:1901.08294 (2019)

    work page internal anchor Pith review Pith/arXiv arXiv 1901

  19. [19]

    3, 999--1045

    Jian Ding and Jiaming Xia, Exponential decay of correlations in the two-dimensional random field I sing model , Inventiones mathematicae 224 (2021), no. 3, 999--1045

    work page 2021

  20. [20]

    Diederik van Engelenburg and Marcin Lis, An elementary proof of phase transition in the planar XY model , Communications in Mathematical Physics 399 (2022), 85--104

    work page 2022

  21. [21]

    u rg Fr \

    J \"u rg Fr \"o hlich, Robert Israel, Elliot H. Lieb, and Barry Simon, Phase transitions and reflection positivity. I . G eneral theory and long range lattice models , Communications in Mathematical Physics 62 (1978), no. 1, 1--34

    work page 1978

  22. [22]

    u rg Fr \

    J \"u rg Fr \"o hlich and Thomas Spencer, The K osterlitz- T houless transition in two-dimensional abelian spin systems and the C oulomb gas , Communications in Mathematical Physics 81 (1981), no. 4, 527--602

    work page 1981

  23. [23]

    Fr\" o hlich, B

    J. Fr\" o hlich, B. Simon, and T. Spencer, Infrared bounds, phase transitions and continuous symmetry breaking, Communications in Mathematical Physics 50 (1976), no. 1, 79--85

    work page 1976

  24. [24]

    Alexander Glazman and Piet Lammers, Delocalisation in the loop O (2) and six-vertex models , In preparation (2023)

    work page 2023

  25. [25]

    3, 1153--1221

    Alexander Glazman and Ioan Manolescu, Uniform L ipschitz functions on the triangular lattice have logarithmic variations , Communications in mathematical physics 381 (2021), no. 3, 1153--1221

    work page 2021

  26. [26]

    53, 2017, pp

    Alessandro Giuliani, Vieri Mastropietro, and Fabio Lucio Toninelli, Height fluctuations in interacting dimers, Annales de l'Institut Henri Poincar \'e , Probabilit \'e s et Statistiques, vol. 53, 2017, pp. 98--168

    work page 2017

  27. [27]

    Alexander Glazman and Ron Peled, On the transition between the disordered and antiferroelectric phases of the 6-vertex model, arXiv preprint arXiv:1909.03436 (2019)

    work page arXiv 1909

  28. [28]

    333, Springer Berlin, Heidelberg, 2006

    Geoffrey R Grimmett, The random-cluster model, Grundlehren der mathematischen Wissenschaften, vol. 333, Springer Berlin, Heidelberg, 2006

    work page 2006

  29. [29]

    Christophe Garban and Avelio Sep \'u lveda, Quantitative bounds on vortex fluctuations in 2d C oulomb gas and maximum of the integer-valued G aussian free field , arXiv preprint arXiv:2012.01400 (2020)

    work page arXiv 2012

  30. [30]

    , Statistical reconstruction of the G aussian free field and KT transition , arXiv preprint arXiv:2002.12284 (2020)

    work page arXiv 2002

  31. [31]

    3, 227--231

    Richard Holley, Remarks on the FKG inequalities , Communications in Mathematical Physics 36 (1974), no. 3, 227--231

    work page 1974

  32. [32]

    Richard Kenyon, Dominos and the G aussian free field , Annals of probability (2001), 1128--1137

    work page 2001

  33. [33]

    Vital Kharash and Ron Peled, The F r\"ohlich- S pencer proof of the B erezinskii- K osterlitz- T houless transition , arXiv preprint arXiv:1711.04720 (2017)

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  34. [34]

    John Michael Kosterlitz and David James Thouless, Ordering, metastability and phase transitions in two-dimensional systems, Journal of Physics C: Solid State Physics 6 (1973), no. 7, 1181

    work page 1973

  35. [35]

    4, 809--838

    Laurin K \"o hler-Schindler and Vincent Tassion, Crossing probabilities for planar percolation, Duke Mathematical Journal 172 (2023), no. 4, 809--838

    work page 2023

  36. [36]

    1, 531--550

    Piet Lammers, Height function delocalisation on cubic planar graphs, Probability Theory and Related Fields 182 (2022), no. 1, 531--550

    work page 2022

  37. [37]

    6, 061101

    Mathieu Lewin, C oulomb and R iesz gases: The known and the unknown , Journal of Mathematical Physics 63 (2022), no. 6, 061101

    work page 2022

  38. [38]

    2, 1181--1205

    Marcin Lis, On delocalization in the six-vertex model, Communications in Mathematical Physics 383 (2021), no. 2, 1181--1205

    work page 2021

  39. [39]

    Piet Lammers and S \'e bastien Ott, Delocalisation and absolute-value- FKG in the solid-on-solid model , arXiv preprint arXiv:2101.05139 (2021)

    work page arXiv 2021

  40. [40]

    32, Cambridge University Press, 1936, pp

    Rudolf Peierls, On I sing's model of ferromagnetism , Mathematical Proceedings of the Cambridge Philosophical Society, vol. 32, Cambridge University Press, 1936, pp. 477--481

    work page 1936

  41. [41]

    3, 1351--1447

    Ron Peled, High-dimensional L ipschitz functions are typically flat , The Annals of Probability 45 (2017), no. 3, 1351--1447

    work page 2017

  42. [42]

    Scott Sheffield, Random surfaces, Ast \'e risque 304 (2005)

    work page 2005

  43. [43]

    Wendelin Werner and Ellen Powell, Lecture notes on the G aussian free field , arXiv preprint arXiv:2004.04720 (2020)

    work page arXiv 2004