Pith. sign in

REVIEW 2 major objections 5 minor 2 cited by

Bi-infinite incipient cluster in high dimensions

T0 review · 2 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Conditioning critical percolation on two far-apart disjoint connections produces a limiting measure, the bi-infinite incipient cluster, singular with respect to both critical percolation and the standard one-arm incipient cluster.

desk verdict Solid construction of a two-armed IIC, but the singularity claim with PIIC is asserted without proof and needs fixing. read the letter →

arxiv 2506.06559 v1 pith:75ZFN7BS submitted 2025-06-06 math.PR

Manuel Cabezas , Alexander Fribergh , Markus Heydenreich , Antal A. Járai This is my paper
classification math.PR MSC 60K3582B4382B27
keywords percolationincipientinfiniteclusterlaceexpansioncriticalhigh-dimensionaltwo-pointfunctionmean-fieldbehaviordiagrammaticbounds
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

At the critical threshold of high-dimensional percolation, the probability that a local event occurs in the cluster of the origin, given that the origin is disjointly connected to two points $x$ and $x'$, is shown to have a limit as $|x|$, $|x'|$, and $|x-x'|$ all diverge. That limit defines a new measure, the bi-infinite incipient cluster $P_{2-IIC}$, which looks locally like critical percolation but almost surely contains two disjoint infinite occupied paths from the origin. The proof works for any independent translation-invariant model in dimension $d>6$ satisfying three explicit assumptions, and it shows that the new measure is singular with respect to both ordinary critical percolation and the standard one-arm incipient infinite cluster. This matters because the two-arm backbone is the part of a critical cluster believed to control the scaling limit of random walk on the cluster.

What carries the argument

The central mechanism is a double lace expansion: the connection $o \leftrightarrow x$ is expanded first, and then the surviving cluster is expanded a second time along the arm toward $x'$, using pivotal bonds and restricted clusters. The resulting coefficients are controlled by a new bookkeeping system of diagrammatic events and generalized diagrams, in which each term is a collection of disjoint occupied paths indexed by a multigraph and is bounded, via the BK inequality, by a product of two-point functions. The load-bearing graph-theoretic tool is the H-reduction: a diagram containing an H-shaped subgraph with unlabelled endpoints of degree at least three is bounded, up to a constant, by the same diagram with the H's crossbar removed. A reduction theorem (Proposition 4.26) shows that every diagram that arises here, with three marked points and all internal vertices of degree three, reduces to one of two small diagrams, so the many expansion coefficients collapse onto a single triangle bound. Assumption (1.5) is exactly the uniform finiteness of the diagram sums that makes the infinite expansion converge.

What would settle it

Take a fixed cylinder event $F$ and compute $\sigma(F;x,x')/(\tau(x)\tau(x'))$ along two different routes as $|x|,|x'|,|x-x'| \to \infty$ in a model satisfying (1.1) and (1.2) but not (1.5); differing limits, or a mismatch with the explicit ratio in (3.14), would show that the double expansion is not convergent and Theorem 1.1 does not extend to that model.

Watch

Extended reading notes

Core claim

The paper's theorem, Theorem 1.1, states that for independent, translation-invariant percolation on $\mathbb{Z}^d$ with $d>6$, under the two-point asymptotics (1.1), the convolution bound (1.2), and the double-lace-expansion diagram bound (1.5), the ratio $P_{p_c}(F \mid \{o \leftrightarrow x\} \circ \{o \leftrightarrow x'\})$ converges for every cylinder event $F$ as $|x|, |x'|, |x-x'| \to \infty$. The limit is written explicitly in (3.14) as a ratio of sums of expansion coefficients, and it extends to a probability measure $P_{2-IIC}$ on the full $\sigma$-field. The measure is concentrated on configurations with two disjoint infinite occupied paths from the origin, locally resembles critical percolation, and is mutually singular with respect to the critical measure and to the earlier incipient infinite cluster measure.

Load-bearing premise

The load-bearing premise is assumption (1.5), which requires a uniform bound on an infinite sum of double-lace-expansion diagrams; it is verified for spread-out percolation with large $L$ and for nearest-neighbor percolation only in sufficiently high dimension, so the theorem does not cover nearest-neighbor percolation in $d=11$ even though the two-point asymptotics (1.1) holds there.

Editorial extensions

If this is right

  • For spread-out percolation in $d>6$ with $L$ large, and for nearest-neighbor percolation in sufficiently high dimension, the two-arm conditional probabilities converge for every cylinder event.
  • The limiting measure $P_{2-IIC}$ is a genuinely new critical object: it has two disjoint infinite paths from the origin and is singular with respect to both $P_{p_c}$ and the one-arm incipient infinite cluster.
  • The explicit formula in (3.14) gives a workable expression for $P_{2-IIC}$ probabilities in terms of the expansion coefficients and the two-point function.
  • Under the same assumptions the normalized two-arm probability $\sigma(x,x') / (\tau(x)\tau(x'))$ converges to a strictly positive constant, confirming the expected mean-field order of the two-arm event.

Reading between the lines

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

  • If assumption (1.5) could be proved from (1.1) alone, the theorem would cover nearest-neighbor percolation in $d=11$; the paper leaves this as its main open question, and the H-reduction machinery suggests the missing step is a sharper diagram estimate rather than a structural change.
  • The same diagrammatic-event and H-reduction framework is a natural template for a three-arm or $k$-arm expansion; a concrete test would be to see whether the reduction theorem still terminates on one or two terminal diagrams for a three-arm event.
  • The paper's motivation implies that if one could prove the expected ergodicity of $P_{2-IIC}$ along the backbone, random-walk scaling limits on critical percolation clusters would follow; that ergodicity is asserted as a belief, not proved here.
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.

Referee Report

2 major / 5 minor

Summary. The paper constructs a new percolation measure, called the bi-infinite incipient cluster P2-IIC, for high-dimensional critical percolation. The construction is by conditioning the origin to be disjointly connected to two far-away points x and x', and then taking the limit as |x|, |x'|, and |x-x'| tend to infinity. Theorem 1.1 asserts that this limit exists for cylinder events under the assumptions (1.1), (1.2), and (1.5), and that it extends to a probability measure. The proof is based on a double lace expansion: Proposition 2.1 expands the two-arm probability into a sum of lace-expansion coefficients plus a remainder, Proposition 3.1 gives diagrammatic bounds on those coefficients, and Section 3.2 derives the limiting ratio. Sections 4 and 5 develop a systematic diagrammatic framework, including the H-reduction method, to prove the coefficient bounds. The paper further claims that P2-IIC is concentrated on configurations with two disjoint infinite occupied paths and is mutually singular with the usual incipient infinite cluster measures.

Significance. If the main theorem is correct, the paper provides a genuinely new IIC-type measure in the mean-field regime, with a construction that is not circular: the assumptions (1.1), (1.2), and (1.5) are external lace-expansion hypotheses with known model-dependent verification, and no fitted parameters appear. The double lace expansion for the second arm and the H-reduction framework for diagrams are substantial methodological contributions that may be useful for studying multi-arm events and k-point functions. The main reservations are that the advertised mutual singularity with the usual IIC measures is not proved anywhere, and that the scope of assumption (1.5) is strictly narrower than the two-point bound (1.1), excluding, for instance, nearest-neighbor percolation in d=11. The paper's value does not depend on the singularity claim for the existence theorem, but the singularity claim is a central advertised novelty and needs either proof or explicit weakening.

major comments (2)
  1. [Abstract and Section 1.3] The mutual singularity assertion is not proved in the manuscript. The paper proves existence of a limit measure, but it never identifies an event that has probability one under P2-IIC and probability zero under the usual PIIC measures. The event 'there exist two disjoint infinite occupied paths from the origin' has P2-IIC-probability one, as is immediate from the defining cylinder-event limits, but no theorem or reference is supplied showing that PIIC assigns probability zero to this event. Since the abstract's claim that P2-IIC is 'mutually singular' with existing IIC measures is a central advertised novelty, this is a load-bearing gap. Please either prove the singularity (for example, by establishing that high-dimensional PIIC has at most one infinite path almost surely, or by finding another distinguishing event) or explicitly weaken the claim to a conjecture or a conditional statement.
  2. [End of Section 3.2, proof of Theorem 1.1] The extension from special cylinder events F00 to a probability measure on the full sigma-field F is only justified by the sentence 'Since F00 is a ∩-stable generator of the sigma algebra F = sigma(F00), this therefore determines the measure P2-IIC on F uniquely.' This is insufficient: F00 is not an algebra, and one must show that the finite-dimensional limiting probabilities are consistent and then invoke a standard extension theorem, or prove countable additivity on the cylinder algebra. Because the existence of a genuine probability measure is part of Theorem 1.1, please supply the missing extension argument.
minor comments (5)
  1. [Lemma 4.25] In the proof of Lemma 4.25, the text says the convolution bound holds 'whenever 3/2(d-2)>d whenever d<6'; this should read 'd>6', which is the regime relevant to the paper.
  2. [Corollary 3.4] In the lower-bound computation, the displayed identity 'E[I{u↔x} tau^{eC(u)}(u',x')] = P(x↔x) P(x'↔x') - ...' appears to contain a typo: the first product should presumably be tau(u,x) tau(u',x') (or P(u↔x) P(u'↔x')), not P(x↔x)P(x'↔x'), both of which equal one.
  3. [Section 1.1 and Theorem 1.1] Assumption (1.5) is substantially stronger than the two-point bound (1.1), and the paper itself notes that it is not known for nearest-neighbor percolation in d=11, where (1.1) holds. The abstract and introduction would benefit from stating this limitation more prominently, since a reader may otherwise infer that the theorem covers all models satisfying (1.1).
  4. [Sections 1.3 and 3.2] The fact that P2-IIC is concentrated on configurations with two disjoint infinite occupied paths is asserted but not explicitly proved. This follows from the conditioning: for the cylinder event F_n that there are two disjoint connections from the origin to distance n, the conditioning event {o↔x} o {o↔x'} implies F_n whenever |x|,|x'|>n, so P2-IIC(F_n)=1 for every n. Stating this as a short lemma would make the paper easier to read.
  5. [Throughout] There are several typographical errors, for example 'probablity' in the description of spread-out percolation and 'there union equals' in the definition of F00; these should be corrected in a revision.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the P2-IIC limit is derived from explicit assumptions via a self-contained double lace expansion; self-citations to [19] are prior-work inputs, not hidden consequences of the theorem.

full rationale

The central claim is conditional on the three stated hypotheses (1.1), (1.2), and (1.5). The proof of Theorem 1.1 does not fit any parameter and does not rename an input as a prediction: the limit defining P2-IIC is evaluated through the double lace expansion identity (3.5), with the numerator and denominator computed in Lemma 3.3 and Corollary 3.4, yielding the explicit formula (3.14). Assumption (1.5) is a genuine diagrammatic convergence condition that is strictly stronger than the two-point bound, and the paper openly records where it is known, citing [19] for spread-out percolation with L large and for nearest-neighbor percolation in sufficiently large dimension. Although [19] shares an author, it is prior published work and is not the target construction, so the citation does not smuggle in the bi-infinite measure or its singularity. Similarly, the remainder bound (3.4) is stated as following from [19, (4.21)-(4.22)]; this is a standard lace-expansion estimate and is not equivalent to the paper's conclusion. The asserted mutual singularity with PIIC is an unproved global claim, but an unsupported or even false claim is a correctness gap, not a circularity, because it is not used as a premise in deriving the existence of the limit. No equation or definition reduces by construction to its own input, and no fitted quantity is relabeled as a prediction.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

The central claim rests on three explicit model-dependent assumptions (1.1), (1.2), (1.5) from the lace expansion literature, which the paper states transparently and discusses in Open Problems 1.6. No free parameters are fitted and no new entities are postulated; the bi-infinite incipient cluster is constructed as a limit, not assumed. The graph-theoretic tools (BK inequality, Menger, Hara-Slade expansion, convolution bounds) are standard external results.

assumptions (7)
  • domain assumption Assumption (1.1): tau(0,x) = A_d |x|^{2-d}(1+o(1)) at criticality.
    Verified for nearest-neighbor d>=11 and spread-out d>6 with large L (Hara, Fitzner-van der Hofstad, Hara-van der Hofstad-Slade); conjectured for NN d>6. Used throughout for convolution bounds and the limit (3.12). Invoked in Section 1.1.
  • domain assumption Assumption (1.2): sum_y D(0,y)tau(y,x) <= C tau(0,x).
    Used to bound the second term in Lemma 3.3 and to control summations over v0; proved for NN and spread-out in Lemma A.1 under decay of D and (1.1).
  • domain assumption Assumption (1.5): convergence of the lace expansion diagram sum sup_{u,v,x} sum_{N>=1} A^{(N)}(u,v,x) / (tau(u,v)tau(u,x)tau(v,x)) < infinity.
    This is the key convergence input for the double lace expansion. It is verified for spread-out d>6 with L large and for NN in sufficiently large dimension (d>d*), not for NN in d=11. The paper notes this gap in Open Problems 1.6 and in Section 1.1.
  • standard math Hara-Slade lace expansion identity (Lemma 2.3) for the pivotal decomposition of tau^S.
    Used in Section 2.1 to expand the first arm; taken from [12].
  • standard math BK inequality for disjoint occurrence.
    Used extensively to bound diagrammatic events by products of two-point functions, cf. [4, 23].
  • standard math Menger's theorem for 3-connectivity and edge cutsets.
    Used in the graph-theoretic proofs of H-reduction (Lemma 4.23, Proposition 4.21).
  • standard math Convolution bounds of Hara-van der Hofstad-Slade [11, Prop 1.7].
    Used for the triangle diagram (3.7) and decay estimates in Section 3.2 and Lemma 4.25.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Bi-infinite incipient cluster in high dimensions." pith.science (2026). https://pith.science/paper/75ZFN7BS

@misc{pith2026250606559,
  author       = {Pith},
  title        = {Pith review of: Bi-infinite incipient cluster in high dimensions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/75ZFN7BS}},
  note         = {Machine review of arXiv:2506.06559}
}
abstract

We consider high-dimensional percolation at the critical threshold. We condition the origin to be disjointly connected to two points, $x$ and $x'$, and subsequently take the limit as $|x|$, $|x'|$ as well as $|x-x'|$ diverge to infinity. This limiting procedure gives rise to a new percolation measure that locally resembles critical percolation but is concentrated on configurations with two disjoint infinite occupied paths. We coin this the bi-infinite incipient percolation cluster. It is mutually singular with respect to incipient infinite clusters that have been constructed in the literature. We achieve the construction through a double lace expansion of the cluster.

Figures

Figures reproduced from arXiv: 2506.06559 by the authors.

Figure 1
Figure 1. Diagrammatic estimates for the proof of the Proposition 4.16 when both x and y have degree larger than 3. coincides with f1 in V (G˜) and coincides with f2 in V (G) \ V (G˜). We can write Diag((G, S, ℓ)) = X f∈LG Y {z,w}∈E(G) τ (f(z), f(w)). = X f1∈LG˜ X f2∈LG\G˜ Y {z,w}∈E(G) τ (f1,2(z), f1,2(w)) = X f1∈LG˜ Y {z,w}∈E∗(G) τ (f1,2(z), f1,2(w)) X f2∈LG\G˜ Y {z,w}∈E(G)\E∗(G) τ (f1,2(z), f1,2(w)), where, in the last step… view at source ↗
Figure 2
Figure 2. Diagrammatic estimates for the proof of the first case of Proposition 4.16. Dotted edges contribute only a constant factor and therefore can be erased. Repeating the strategy on each of the terms we get that the display above is bounded by Cτ (f1(z), f1(w))τ (f1(s), f1(t))h X f2(x),f2(y)∈Zd τ (f2(x), f1(w))τ (f2(y), f1(t))τ (f2(x), f2(y))+ + X f2(x),f2(y)∈Zd τ (f2(x), f1(w))τ (f1(s), f2(y))τ (f2(x), f2(y))+ + X f2(x… view at source ↗
Figure 3
Figure 3. Diagrammatic estimates for the proof of the second sub-case of Propo￾sition 4.16. Dotted edges contribute only a constant factor and therefore can be erased. □ [PITH_FULL_IMAGE:figures/full_fig_p034_3.png] view at source ↗

Discussion (0). Sign in 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

  1. Super-Brownian limits and the $k$-point function for high-dimensional percolation

    math.PR 2026-07 accept novelty 8.0 of 10

    High-dimensional critical percolation clusters rescale to super-Brownian excursion, verifying the 1984 Aizenman–Newman k-point conjecture under lace-expansion hypotheses.

  2. On Loops in critical high-dimensional percolation

    math.PR 2025-07 conditional novelty 8.0 of 10

    Critical percolation in high dimensions contains a tight number of macroscopic loop-clusters, each with essentially one large loop, with subsequential Hausdorff scaling limits.

Reference graph

Works this paper leans on

24 extracted references · 22 canonical work pages · cited by 2 Pith papers

  1. [1]

    Ben Arous, M

    G. Ben Arous, M. Cabezas, and A. Fribergh. Scaling limit for the ant in a simple high-dimensional labyrinth. Probab. Theory Related Fields, 174(1-2):553–646, 2019

    work page 2019

  2. [2]

    Ben Arous, M

    G. Ben Arous, M. Cabezas, and A. Fribergh. Scaling limit for the ant in high-dimensional labyrinths. Comm. Pure Appl. Math. , 72(4):669–763, 2019

    work page 2019

  3. [3]

    Scaling limit for the random walk on critical lattice trees

    G. Ben Arous, M. Cabezas, and A. Fribergh. Scaling limit for the random walk on critical lattice trees. Preprint arXiv:2503.22538 [math.PR], 2025

    work page Pith review arXiv 2025

  4. [4]

    van den Berg and H

    J. van den Berg and H. Kesten. Inequalities with applications to percolation and reliability. J. Appl. Probab., 22(3):556–569, 1985

    work page 1985

  5. [5]

    Cabezas, A

    M. Cabezas, A. Fribergh, M. Holmes, and E. Perkins. Random skeletons in high-dimensional lattice trees. Preprint arXiv:2503.19230 [math.PR], 2025

    arXiv 2025

  6. [6]

    Chatterjee, P

    S. Chatterjee, P. Chinmay, J. Hanson, and P. Sosoe. Robust construction of the incipient infinite cluster in high dimensional critical percolation. Preprint arXiv:2502.10882 [math.PR], 2025

    arXiv 2025

  7. [7]

    Damron and A

    M. Damron and A. Sapozhnikov. Outlets of 2d invasion percolation and multiple-armed incipient infinite clusters. Probab. Theory Relat. Fields , 150(1-2):257–294, 2011

    work page 2011

  8. [8]

    Fitzner and R

    R. Fitzner and R. van der Hofstad. Mean-field behavior for nearest-neighbor percolation in d >10. Electron. J. Probab., 22:65, 2017. Id/No 43

    work page 2017

Show all 24 references
  1. [9]

    Grimmett

    G. Grimmett. Percolation, volume 321 of Grundlehren der Mathematischen Wissenschaften [Fundamental Principles of Mathematical Sciences] . Springer-Verlag, Berlin, second edition, 1999

    1999

  2. [10]

    T. Hara. Decay of correlations in nearest-neighbour self-avoiding walk, percolation, lattice trees and animals. Ann. Probab., 36(2):530–593, 2008. 70 MANUEL CABEZAS, ALEXANDER FRIBERGH, MARKUS HEYDENREICH, AND ANTAL A. J ´ARAI

    2008

  3. [11]

    T. Hara, R. van der Hofstad, and G. Slade. Critical two-point functions and the lace expansion for spread-out high-dimensional percolation and related models. Ann. Probab., 31(1):349–408, 2003

    2003

  4. [12]

    Hara and G

    T. Hara and G. Slade. Mean-field critical behaviour for percolation in high dimensions. Comm. Math. Phys. , 128(2):333–391, 1990

    1990

  5. [13]

    Hara and G

    T. Hara and G. Slade. The self-avoiding-walk and percolation critical points in high dimensions. Combin. Probab. Comput., 4(3):197–215, 1995

    1995

  6. [14]

    Hara and G

    T. Hara and G. Slade. The scaling limit of the incipient infinite cluster in high-dimensional percolation. I. Critical exponents. J. Statist. Phys. , 99(5-6):1075–1168, 2000

    2000

  7. [15]

    Hara and G

    T. Hara and G. Slade. The scaling limit of the incipient infinite cluster in high-dimensional percolation. II. Integrated super-Brownian excursion. J. Math. Phys. , 41(3):1244–1293, 2000

    2000

  8. [16]

    Heydenreich, R

    M. Heydenreich, R. van der Hofstad, and W. J. T. Hulshof. High-dimensional incipient infinite clusters revisited. J. Stat. Phys. , 155(5):966–1025, 2014

    2014

  9. [17]

    Heydenreich, R

    M. Heydenreich, R. van der Hofstad, and A. Sakai. Mean-field behavior for long- and finite range Ising model, percolation and self-avoiding walk. J. Stat. Phys. , 132(6):1001–1049, 2008

    2008

  10. [18]

    Heydenreich and R

    M. Heydenreich and R. van der Hofstad. Progress in high-dimensional percolation and random graphs . CRM Short Courses. Springer, Cham; Centre de Recherches Math´ ematiques, Montreal, QC, 2017

    2017

  11. [19]

    van der Hofstad and A

    R. van der Hofstad and A. A. J´ arai. The incipient infinite cluster for high-dimensional unoriented percolation. J. Statist. Phys. , 114(3-4):625–663, 2004

    2004

  12. [20]

    A. A. J´ arai. Incipient infinite percolation clusters in 2D. Ann. Probab., 31(1):444–485, 2003

    2003

  13. [21]

    H. Kesten. The incipient infinite cluster in two-dimensional percolation. Probab. Theory Related Fields , 73(3):369–394, 1986

    1986

  14. [22]

    Markering

    M. Markering. Two-sided infinite self-avoiding walk in high dimensions. Preprint arXiv:2410.01507 [math.PR], 2024

    2024 arXiv

  15. [23]

    G. Slade. The Lace Expansion and its Applications , volume 1879 of Lecture Notes in Mathematics . Springer- Verlag, Berlin, 2006

    2006

  16. [24]

    C.-L. Yao. Multi-arm incipient infinite clusters in 2d: scaling limits and winding numbers. Ann. Inst. Henri Poincar´ e, Probab. Stat., 54(4):1848–1876, 2018. Manuel Cabezas, Pontificia Universidad Cat ´olica de Chile, F acultad de Matem´aticas, Campus San Joaqu´ın, A venida V...

    2018

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.