arxiv: 2605.09459 · v1 · submitted 2026-05-10 · 🧮 math.PR

Recognition: 2 theorem links

· Lean Theorem

Multipoint connectivity in the branching interlacement process

Louis Vanhaelewyn

Pith reviewed 2026-05-12 04:50 UTC · model grok-4.3

classification 🧮 math.PR

keywords branching interlacementmultipoint connectivitytrajectoriessharp boundsrandom walksconnectivity structure

0 comments

The pith

Any two points in the branching interlacement are connected by at most ceiling of d over 4 trajectories, and the bound is sharp almost surely.

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

The paper establishes sharp upper bounds on the minimal number of trajectories from the branching interlacement needed to connect multiple points. For any two points the bound is ceiling of d over 4, and there exist pairs that require exactly this number almost surely. The result generalizes to k points, which connect via at most ceiling of d times (k minus 1) over 4 minus (k minus 2) trajectories, again with the bound achieved almost surely. A sympathetic reader would care because these bounds describe the sparse connectivity structure of a random object that models branching random walks.

Core claim

We show that two points of the interlacement are connected via at most ceiling d/4 trajectories of the interlacement. This upper bound is sharp, in the sense that almost surely there exist two points not connected by ceiling d/4 minus 1 trajectories. We extend this result by proving that k points of the interlacement are connected via at most ceiling d(k-1)/4 minus (k-2) trajectories, and that this bound is also sharp.

What carries the argument

The minimal number of interlacement trajectories required to connect k points in the branching interlacement model.

Load-bearing premise

The branching interlacement is defined as introduced by Zhu and the underlying probability space allows the almost-sure sharpness statements to hold.

What would settle it

An explicit pair of points in the model whose connecting trajectories exceed ceiling of d over 4 in number would falsify the upper bound.

Figures

Figures reproduced from arXiv: 2605.09459 by Louis Vanhaelewyn.

**Figure 1.** Figure 1: Labeling of an infinite invariant tree. For an infinite critical tree, denote by (ξi)i⩾1 the vertices of the spine ordered by a depth-first search from the origin. Then for a vertex u in T we introduce the notation nu, for the index of the node ξnu on the spine that is the nearest vertex from u located on the spine (see [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗

**Figure 2.** Figure 2: Example of a spine decomposition and nearest vertex on the spine ( [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

**Figure 3.** Figure 3: Example of a tree T and the generated tree T⟨7, 9, 12, 15, 16⟩ 7 [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗

**Figure 4.** Figure 4: Example of a set Φ(X, R) for a single random walk X. We only draw the spine of the branching random walk for simplicity. The following lemma provides an upper bound on the number of vertices of a branching random walk which fall in a ball of radius R. We need this result in order to control the capacity of the set Φ for N independent branching random walks. Lemma 20. Let R > 0, and let X be a random walk o… view at source ↗

**Figure 5.** Figure 5: Example of a set Ψ(ω, A, R). Lemma 22. Let σu be a random point measure with distribution Poi(u, W∗ ). Then there exists c > 0, such that for all finite subsets A of Z d , and for all positive R ⩾ 1, we have E h BCap(Ψ(σu, A, R))i ⩾ c min{uBCap(A)R 4 , Rd−4 }. Proof. The independence between NA = NA(σu) and (Xi)1⩽i⩽NA , and Lemma 21 imply that E h BCap(Ψ(σu, A, R))i ⩾ cE h min(NAR 4 , Rd−4 ) i . Recall tha… view at source ↗

**Figure 6.** Figure 6: Example of the construction of A(s) (r, R). Lemma 24. Let s be a positive integer. There exists a finite constant Cs = C(u, d, s) such that for all positive r < R, and x ∈ B(R), (i) Ex h BCap A(s) (r, R) i ⩽ CsRmin(d−4,4s) , (ii) Ex h BCap A(s) (r, R) 2 i ⩽ CsR2 min(d−4,4s) . Proof of (i). We need to show that Ex h BCap A(s) (r, R) i is bounded by both Rd−4 and R4s up to a multiplicative constant. On … view at source ↗

**Figure 7.** Figure 7: Example of a connection of n visible sets. Proof of Lemma 32. For simplicity, we will write Ai for A (ki) i (r, R) if 1 ⩽ i ⩽ n − 1, and An for A (1) n (r, R). Let E = ∃ γ ∈ σ u r,2R : γ connects A1, . . . , An [PITH_FULL_IMAGE:figures/full_fig_p027_7.png] view at source ↗

**Figure 8.** Figure 8: Sketch of proof of the upper bound in Theorem 4 [PITH_FULL_IMAGE:figures/full_fig_p028_8.png] view at source ↗

**Figure 9.** Figure 9: An example of a cut operation, where the weights of the two initial edges are [PITH_FULL_IMAGE:figures/full_fig_p032_9.png] view at source ↗

**Figure 10.** Figure 10: An example of a merge operation, where the weights of the two initial edges are [PITH_FULL_IMAGE:figures/full_fig_p033_10.png] view at source ↗

read the original abstract

We consider the branching interlacement model introduced by Zhu as an analog of Sznitman's random interlacement for branching random walks. We show that two points of the interlacement are connected via at most $\lceil d/4 \rceil$ trajectories of the interlacement, using a different proof than Procaccia and Zhang. This upper bound is sharp, in the sense that almost surely there exist two points not connected by $\lceil d/4\rceil - 1$ trajectories. We extend this result by proving that $k$ points of the interlacement are connected via at most $\lceil d(k-1)/4\rceil -(k-2)$ trajectories, and that this bound is also sharp.

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.

Desk Editor's Note private letter to a colleague

The paper gives a clean multipoint extension of connectivity bounds in branching interlacements with almost-sure sharpness, using a different proof for the base case.

read the letter

The main point is that this work extends the two-point connectivity result for branching interlacements to arbitrary k points and supplies explicit sharp bounds. For two points it recovers an upper bound of ceil(d/4) trajectories with a new argument, and for k points the bound becomes ceil(d(k-1)/4) minus (k-2), again shown to be tight almost surely. That formula and the alternative proof are the concrete additions relative to the cited earlier papers on the model from Zhu and from Procaccia-Zhang. The almost-sure sharpness statements are stated directly on the underlying probability space, which keeps the argument from looking circular or fitted. The paper builds on the existing construction without introducing new free parameters or invented objects, and the citation pattern is standard for this corner of probability. The central claims therefore look internally consistent once the branching interlacement model is granted. A soft spot is that the result stays narrowly inside percolation-type questions for this specific random-media model; the dimension dependence and the handling of the branching mechanism for larger k are not immediately visible without the full derivations, though the stress-test note finds no obvious gap. The work is aimed at specialists already familiar with interlacements and branching random walks. A reader working on multipoint connectivity or quantitative percolation bounds in random media will get a usable new formula and a sharpness statement that can be checked against the model. It is worth sending to a serious referee because the extension is explicit, the claims are falsifiable in principle, and the alternative proof may be of independent interest even if the overall impact remains inside math.PR.

Referee Report

0 major / 2 minor

Summary. The manuscript studies multipoint connectivity in the branching interlacement process introduced by Zhu as an analog of Sznitman's random interlacements for branching random walks. It proves that any two points of the interlacement are connected by at most ⌈d/4⌉ trajectories of the interlacement, using a proof technique different from that of Procaccia and Zhang; this upper bound is sharp in the sense that almost surely there exist two points not connected by ⌈d/4⌉−1 trajectories. The result is extended to k points, which are connected by at most ⌈d(k−1)/4⌉−(k−2) trajectories, with the bound also shown to be sharp almost surely.

Significance. If the proofs hold, the paper delivers sharp, quantitative bounds on the trajectory connectivity structure of the branching interlacement, including an almost-sure sharpness statement and a natural multipoint generalization. These strengthen the understanding of percolation-type properties in branching random media and the alternative proof method relative to prior work is a constructive contribution.

minor comments (2)

The abstract does not state the range of dimensions d for which the results are claimed; adding this (presumably d ≥ 5 or similar, given the branching mechanism) would improve clarity for readers.
In the multipoint bound ⌈d(k−1)/4⌉−(k−2), a short remark explaining the origin of the −(k−2) correction term (e.g., how it arises inductively from the two-point case) would help readers follow the extension without consulting the full proof.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their positive summary of our work on multipoint connectivity in the branching interlacement process, for recognizing the significance of the sharp quantitative bounds and the alternative proof technique, and for recommending minor revision. No specific major comments appear in the report.

Circularity Check

0 steps flagged

No significant circularity; derivation is self-contained

full rationale

The paper establishes connectivity bounds for the branching interlacement model via direct probabilistic arguments and almost-sure sharpness statements. The model is taken as given from Zhu's prior construction (an external reference, not a self-citation by the present author), and the upper/lower bounds on the number of trajectories connecting points are proved independently rather than fitted or defined in terms of the target quantities. No equations reduce by construction to inputs, no uniqueness theorems are imported from the same authors, and the extension to k points follows from the same proof technique without circular renaming or ansatz smuggling. The result is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Only the abstract is available; no explicit free parameters, axioms, or invented entities can be identified from the given text.

pith-pipeline@v0.9.0 · 5407 in / 1103 out tokens · 33485 ms · 2026-05-12T04:50:07.833478+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

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

IndisputableMonolith/Foundation/AlexanderDuality.lean alexander_duality_circle_linking unclear
We show that two points of the interlacement are connected via at most ⌈d/4⌉ trajectories... k points ... ⌈d(k−1)/4⌉−(k−2) trajectories
IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear
G(x,y)≍1/(1+∥x−y∥^{d−4}) ... BCap(B(R))≍R^{d−4}

Reference graph

Works this paper leans on

7 extracted references · 7 canonical work pages

[1]

[Ald91] D. Aldous. Asymptotic fringe distributions for general families of random trees. Ann. Appl. Probab.1.2 (1991), pp. 228–266. [ASS23] A. Asselah, B. Schapira, and P. Sousi. Local times and capacity for transient branching random walks

work page 1991
[2]

[BW22] T

arXiv:2303.17572. [BW22] T. Bai and Y. Wan. Capacity of the range of tree-indexed random walk.Ann. Appl. Probab.32.3 (2022), pp. 1557–1589. [ˇCP12] J. ˇCern´ y and S. Popov. On the internal distance in the interlacement set.Elec- tron. J. Probab17.29 (2012), pp. 1–25. [DGRST24] H. Duminil-Copin, S. Goswami, P.-F. Rodriguez, F. Severo, and A. Teixeira. A c...

work page arXiv 2022
[3]

Erd¨ os and S

[ET60] P. Erd¨ os and S. J. Taylor. Some intersection properties of random walk paths. Acta Mathematica Academiae Scientiarum Hungarica11 (1960), pp. 231–248. [HHS03] 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 (2003), pp. 349...

work page 1960
[4]

[Szn10] A.-S

arXiv:2504.07525. [Szn10] A.-S. Sznitman. Vacant Set of Random Interlacements and Percolation.Ann. Math.171 (2010), pp. 2039–2087. [Szn12] A.-S. Sznitman. Random interlacements and the Gaussian free field.The Annals of Probability40 (Feb. 2012). [Zhu16a] Q. Zhu. On the critical branching random walk I: Branching capacity and visiting probability

work page arXiv 2010
[5]

[Zhu16b] Q

arXiv:1611.10324. [Zhu16b] Q. Zhu. On the critical branching random walk II: Branching capacity and branching recurrence

work page arXiv
[6]

[Zhu18] Q

arXiv:1612.00161. [Zhu18] Q. Zhu. Branching interlacements and tree-indexed random walks in tori

work page arXiv
[7]

arXiv:1812.10858. 38

work page arXiv