Generalized intersection exponents and local cut points for three-dimensional Brownian loop soup

Ruixuan Li; Xinyi Li; Yifan Gao

arxiv: 2605.17494 · v1 · pith:KEPWWYGInew · submitted 2026-05-17 · 🧮 math.PR

Generalized intersection exponents and local cut points for three-dimensional Brownian loop soup

Yifan Gao , Ruixuan Li , Xinyi Li This is my paper

Pith reviewed 2026-05-19 22:41 UTC · model grok-4.3

classification 🧮 math.PR

keywords Brownian loop soupgeneralized intersection exponentslocal cut pointsHausdorff dimensionintersection exponentsBrownian motionnon-intersection probabilities

0 comments

The pith

Generalized intersection exponents exist for the three-dimensional Brownian loop soup and relate to the dimension of its local cut points.

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

The paper seeks to extend the concept of intersection exponents from plain Brownian motion to the Brownian loop soup in three dimensions at low intensities. These generalized exponents describe the rate at which probabilities of certain non-intersection events decay. By establishing a separation lemma specific to the soup, the authors obtain bounds that allow defining the exponents. They connect these exponents to the Hausdorff dimension of local cut points in the soup. The continuity of the exponents at zero intensity shows that the cut-point dimension exceeds one even when the soup has only a small density of loops.

Core claim

We establish the existence of generalized intersection exponents (GIE) and prove an up-to-constants estimate for these probabilities by means of a separation lemma tailored to this setting. We also relate the Hausdorff dimension of the set of local cut points of the three-dimensional Brownian loop soup to the GIE, and show that the GIE is continuous at intensity zero, where it reduces to the classical Brownian intersection exponent. In particular, this implies that, for sufficiently small intensity parameters, the set of local cut points has Hausdorff dimension strictly larger than 1.

What carries the argument

Generalized intersection exponents (GIE) defined via non-intersection probabilities for the loop soup, which serve as the link to the Hausdorff dimension of local cut points.

If this is right

The non-intersection probabilities admit exponential bounds controlled by the GIE.
The Hausdorff dimension of local cut points is a function of the GIE.
The GIE converge to the classical Brownian intersection exponent as intensity tends to zero.
For small positive intensities the local cut points have Hausdorff dimension strictly greater than one.

Where Pith is reading between the lines

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

This continuity suggests that adding a small loop soup perturbs the intersection properties of Brownian paths in a controllable way.
Similar generalized exponents could be defined for other random collections of paths in three dimensions.
The result on dimension greater than one may indicate that local cut points affect global connectivity features of the soup.

Load-bearing premise

The tailored separation lemma for the loop soup provides the up-to-constants estimate required to establish the existence of the generalized intersection exponents.

What would settle it

A demonstration that the non-intersection probabilities do not decay at a rate consistent with a positive generalized intersection exponent for small intensities would falsify the main claims.

read the original abstract

We study generalized non-intersection probabilities for the three-dimensional Brownian loop soup at subcritical intensities. We establish the existence of generalized intersection exponents (GIE) and prove an up-to-constants estimate for these probabilities by means of a separation lemma tailored to this setting. We also relate the Hausdorff dimension of the set of local cut points of the three-dimensional Brownian loop soup to the GIE, and show that the GIE is continuous at intensity zero, where it reduces to the classical Brownian intersection exponent. In particular, this implies that, for sufficiently small intensity parameters, the set of local cut points has Hausdorff dimension strictly larger than $1$.

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 defines generalized intersection exponents for 3D Brownian loop soups via non-intersection probabilities, proves their existence and continuity at zero intensity using a tailored separation lemma, and derives that local cut points have Hausdorff dimension strictly above 1 for small intensities.

read the letter

The core advance is a workable definition of generalized intersection exponents for the loop soup model together with an up-to-constants bound that lets them pass to a limit. They then tie those exponents to the dimension of local cut points and recover the classical Brownian exponent in the zero-intensity limit. That last step is clean and gives the strict inequality dim > 1 for small positive intensity, which is the concrete geometric output. The separation lemma is the technical engine; it has to deliver scale-invariant control that stays uniform as the intensity parameter approaches zero, and the abstract indicates they adapted it to the Poisson point process of loops. If the constants really are independent of intensity near zero, the continuity and dimension claims follow directly. The argument stays inside the existing program on random fractals and intersection exponents, so the citations line up with the classical references rather than circling back on themselves. The main soft spot is that everything rests on the details of that lemma. Any hidden dependence on intensity in the constants would break the continuity statement and weaken the dimension formula. Since the work is purely analytic with no numerical checks, a referee would need to verify the uniformity claim line by line. This is aimed at people already working on Brownian motion, loop soups, or Hausdorff dimensions of random sets in three dimensions. A reader who knows the classical intersection-exponent literature will see exactly where the new objects sit and what the geometric payoff is. It is worth sending to a serious referee; the claims are specific enough and the reduction to the zero-intensity case is a good anchor, so the paper should get a careful read rather than a desk rejection.

Referee Report

1 major / 2 minor

Summary. The paper defines generalized non-intersection probabilities for the three-dimensional Brownian loop soup at subcritical intensities. It establishes existence of the generalized intersection exponents (GIE) via an up-to-constants estimate obtained from a separation lemma adapted to the Poissonian loop soup, relates the Hausdorff dimension of the local cut-point set to the GIE, and proves that the GIE is continuous at intensity zero, recovering the classical Brownian intersection exponent. As a consequence, the local cut points have Hausdorff dimension strictly larger than 1 for all sufficiently small intensities.

Significance. If the separation lemma supplies the claimed uniform bounds, the work supplies the first rigorous extension of intersection-exponent techniques to the loop-soup setting and yields a concrete dimension formula for local cut points that is new even for small intensities. The continuity statement at zero intensity is a natural and useful bridge to the classical theory.

major comments (1)

[separation lemma (presumably §3 or §4)] The existence of the GIE, the dimension formula for local cut points, and the continuity statement at intensity zero all rest on the separation lemma delivering two-sided bounds that are uniform in the intensity parameter near zero. The manuscript should state the lemma explicitly (including the precise form of the constants) and verify uniformity; without this the limit defining the GIE may fail to exist as a positive finite number.

minor comments (2)

Notation for the intensity parameter and the loop-soup measure should be introduced once and used consistently; several passages in the abstract and introduction repeat the same definition.
The statement that the dimension is 'strictly larger than 1' for small intensities should be accompanied by an explicit lower bound on the intensity interval on which the inequality holds.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the careful reading and constructive feedback on our manuscript. We address the major comment below and will incorporate the suggested clarifications in the revised version.

read point-by-point responses

Referee: [separation lemma (presumably §3 or §4)] The existence of the GIE, the dimension formula for local cut points, and the continuity statement at intensity zero all rest on the separation lemma delivering two-sided bounds that are uniform in the intensity parameter near zero. The manuscript should state the lemma explicitly (including the precise form of the constants) and verify uniformity; without this the limit defining the GIE may fail to exist as a positive finite number.

Authors: We agree that greater explicitness will strengthen the presentation. In the revised manuscript we will isolate the separation lemma as a standalone statement in Section 3, writing the two-sided bounds with explicit constants that are independent of the intensity parameter for all intensities in a fixed neighborhood of zero. A short additional argument will verify uniformity by exploiting the Poissonian character of the loop soup: the probability of a loop crossing a separating surface is controlled by a first-moment calculation whose intensity factor remains bounded as the intensity tends to zero, while the complementary lower bound follows from a second-moment estimate that likewise stays uniform. These controls ensure that the up-to-constants estimate for the generalized non-intersection probabilities is uniform near zero intensity, so that the limit defining each GIE exists and is positive and finite. The same uniform bounds are then used without change for the dimension formula of local cut points and for the continuity statement at intensity zero. revision: yes

Circularity Check

0 steps flagged

No circularity: GIE defined from non-intersection probabilities and proved via independent separation lemma

full rationale

The paper defines generalized intersection exponents directly from limits of non-intersection probabilities for the 3D Brownian loop soup at subcritical intensities. Existence, up-to-constants bounds, continuity at intensity zero (reducing to the classical exponent), and the Hausdorff dimension relation for local cut points are all derived from a separation lemma adapted to the Poisson point process of loops. No quoted step reduces any claimed result to a fitted parameter, self-citation chain, or definitional tautology; the lemma supplies the required scale-invariant estimates without presupposing the target exponents or dimension formula. The derivation is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Only the abstract is available, so the ledger is necessarily incomplete. The work appears to rest on the standard construction of the Brownian loop soup, the existence of a separation lemma for the relevant probabilities, and the usual properties of Hausdorff dimension in metric spaces.

axioms (2)

domain assumption The three-dimensional Brownian loop soup at subcritical intensities is well-defined and satisfies the usual Markov and scaling properties.
Invoked implicitly when defining non-intersection probabilities and local cut points.
ad hoc to paper A separation lemma tailored to the loop-soup setting supplies two-sided bounds on the generalized non-intersection probabilities.
Explicitly mentioned as the tool used to prove the up-to-constants estimate for the GIE.

pith-pipeline@v0.9.0 · 5635 in / 1585 out tokens · 28837 ms · 2026-05-19T22:41:39.804603+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

?

unclear
Relation between the paper passage and the cited Recognition theorem.

We establish the existence of generalized intersection exponents (GIE) and prove an up-to-constants estimate for these probabilities by means of a separation lemma tailored to this setting. ... dimH(Gloc) = max{2 − ξ_α(1,1), 0}
IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

Theorem 1.1. ... C1 e^{-r ξ_α(k,λ)} ≤ p(α,k,r,λ) ≤ C2 e^{-r ξ_α(k,λ)}

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.

Reference graph

Works this paper leans on

31 extracted references · 31 canonical work pages

[1]

Burdzy and G

K. Burdzy and G. F. Lawler. Nonintersection exponents for Brownian paths. II. Estimates and applications to a random fractal. Ann. Probab., 18(3):981–1009, 1990

work page 1990
[2]

Cai and J

Z. Cai and J. Ding. One-arm probabilities for metric graph Gaussian free ﬁelds below and at the critical dimension. arXiv preprint arXiv:2406.02397 , 2024

work page arXiv 2024
[3]

Cai and J

Z. Cai and J. Ding. Heterochromatic two-arm probabilities for metric graph Gaussian free ﬁelds. arXiv preprint arXiv:2510.20492 , 2025

work page arXiv 2025
[4]

Cai and J

Z. Cai and J. Ding. On the gap between cluster dimensions of loop soups on R3 and the metric graph of Z3. arXiv preprint arXiv:2510.20526 , 2025

work page arXiv 2025
[5]

Cai and J

Z. Cai and J. Ding. Separation and cut edge in macroscopic clusters for metric graph Gaussian free ﬁelds. arXiv preprint arXiv:2510.20516 , 2025

work page arXiv 2025
[6]

Chang, H

Y. Chang, H. Du, and X. Li. Percolation threshold for metric graph loop soup. Bernoulli, 30(4):3324–3333, 2024

work page 2024
[7]

Chang and A

Y. Chang and A. Sapozhnikov. Phase transition in loop percolation. Probab. Theory Relat. Fields, 164:979–1025, 2016

work page 2016
[8]

J. T. Chayes, L. Chayes, and R. Durrett. Connectivity properties of Mandelbrot’s percolation process. Probab. Theory Relat. Fields , 77(3):307–324, 1988

work page 1988
[9]

J. T. Chayes, L. Chayes, E. Grannan, and G. Swindle. Phase transitions in Mandelbrot’s percolation process in three dimensions. Probab. Theory Relat. Fields , 90(3):291–300, 1991

work page 1991
[10]

Drewitz, A

A. Drewitz, A. Prévost, and P.-F. Rodriguez. Critical one-arm probability for the metric Gaussian free ﬁeld in low dimensions. Probab. Theory Relat. Fields , 2025

work page 2025
[11]

Y. Gao, X. Li, Y. Li, R. Liu, and X. Liu. A boundary Harnack principle and its application to analyticity of 3D Brownian intersection exponents. arXiv preprint arXiv:2411.14921 , 2024

work page arXiv 2024
[12]

Y. Gao, X. Li, and W. Qian. Multiple points on the boundaries of Brownian loop-soup clusters. Ann. Probab. 54(1): 216–268. , 2026. 22

work page 2026
[13]

Garban, G

C. Garban, G. Pete, and O. Schramm. Pivotal, cluster, and interface measures for critical planar percolation. J. Amer. Math. Soc. , 26:939–1024, 2013

work page 2013
[14]

Holden, G

N. Holden, G. Lawler, X. Li, and X. Sun. Minkowski content of Brownian cut points. Ann. Inst. H. Poincaré Probab. Statist. , 58:455–488, 2022

work page 2022
[15]

S. Janson. Bounds on the distributions of extremal values of a scanning process. Stochastic Process. Appl., 18(2):313–328, 1984

work page 1984
[16]

Jego and T

A. Jego and T. Lupu. Three-dimensional Brownian loop soup clusters. arXiv preprint arxiv:2601.04840, 2026

work page arXiv 2026
[17]

G. Lawler. Strict concavity of the intersection exponent for Brownian motion in two and three dimensions. Math. Phys. Electron. J. , 4:paper no. 5., 1998

work page 1998
[18]

Lawler, O

G. Lawler, O. Schramm, and W. Werner. Conformal restriction: the chordal case. J. Amer. Math. Soc. , 16(4):917–955, 2003

work page 2003
[19]

Lawler and W

G. Lawler and W. Werner. The Brownian loop soup. Probab. Theory Relat. Fields , 128:565– 588, 2004

work page 2004
[20]

G. F. Lawler. Hausdorﬀ dimension of cut points for Brownian motion. Electron. J. Probab. , 1:1–20, 1996

work page 1996
[21]

G. F. Lawler. The inﬁnite two-sided loop-erased random walk. Electron. J. Probab. , 25:1–42, 2020

work page 2020
[22]

G. F. Lawler, O. Schramm, and W. Werner. Values of Brownian intersection exponents. I. Half-plane exponents. Acta Math. , 187(2):237–273, 2001

work page 2001
[23]

G. F. Lawler, O. Schramm, and W. Werner. Values of Brownian intersection exponents. II. Plane exponents. Acta Math. , 187(2):275–308, 2001

work page 2001
[24]

G. F. Lawler, O. Schramm, and W. Werner. Analyticity of intersection exponents for planar Brownian motion. Acta Math. , 189(2):179–201, 2002

work page 2002
[25]

G. F. Lawler, O. Schramm, and W. Werner. Values of Brownian intersection exponents. III. Two-sided exponents. Ann. Inst. H. Poincaré Probab. Statist. , 38(1):109–123, 2002

work page 2002
[26]

G. F. Lawler and B. Vermesi. Fast convergence to an invariant measure for non-intersecting 3-dimensional Brownian paths. ALEA Lat. Am. J. Probab. Math. Stat. , 9(2):717–738, 2012

work page 2012
[27]

W. Qian. Generalized disconnection exponents. Probab. Theory Relat. Fields , 179(1):117–164, 2021

work page 2021
[28]

Sapozhnikov and D

A. Sapozhnikov and D. Shiraishi. On Brownian motion, simple paths, and loops. Probab. Theory Relat. Fields , 172(3):615–662, 2018

work page 2018
[29]

Sheﬃeld and W

S. Sheﬃeld and W. Werner. Conformal loop ensembles: the Markovian characterization and the loop-soup construction. Annals of Mathematics , 176:1827–1917, 2012. 23

work page 1917
[30]

W. Werner. On clusters of Brownian loops in d dimensions. In In and Out of Equilibrium 3: Celebrating Vladas Sidoravicius , pages 797–817. Springer, 2020

work page 2020
[31]

W. Werner. A switching identity for cable-graph loop soups and Gaussian free ﬁelds. arXiv preprint arXiv:2502.06754 , 2025. 24

work page arXiv 2025

[1] [1]

Burdzy and G

K. Burdzy and G. F. Lawler. Nonintersection exponents for Brownian paths. II. Estimates and applications to a random fractal. Ann. Probab., 18(3):981–1009, 1990

work page 1990

[2] [2]

Cai and J

Z. Cai and J. Ding. One-arm probabilities for metric graph Gaussian free ﬁelds below and at the critical dimension. arXiv preprint arXiv:2406.02397 , 2024

work page arXiv 2024

[3] [3]

Cai and J

Z. Cai and J. Ding. Heterochromatic two-arm probabilities for metric graph Gaussian free ﬁelds. arXiv preprint arXiv:2510.20492 , 2025

work page arXiv 2025

[4] [4]

Cai and J

Z. Cai and J. Ding. On the gap between cluster dimensions of loop soups on R3 and the metric graph of Z3. arXiv preprint arXiv:2510.20526 , 2025

work page arXiv 2025

[5] [5]

Cai and J

Z. Cai and J. Ding. Separation and cut edge in macroscopic clusters for metric graph Gaussian free ﬁelds. arXiv preprint arXiv:2510.20516 , 2025

work page arXiv 2025

[6] [6]

Chang, H

Y. Chang, H. Du, and X. Li. Percolation threshold for metric graph loop soup. Bernoulli, 30(4):3324–3333, 2024

work page 2024

[7] [7]

Chang and A

Y. Chang and A. Sapozhnikov. Phase transition in loop percolation. Probab. Theory Relat. Fields, 164:979–1025, 2016

work page 2016

[8] [8]

J. T. Chayes, L. Chayes, and R. Durrett. Connectivity properties of Mandelbrot’s percolation process. Probab. Theory Relat. Fields , 77(3):307–324, 1988

work page 1988

[9] [9]

J. T. Chayes, L. Chayes, E. Grannan, and G. Swindle. Phase transitions in Mandelbrot’s percolation process in three dimensions. Probab. Theory Relat. Fields , 90(3):291–300, 1991

work page 1991

[10] [10]

Drewitz, A

A. Drewitz, A. Prévost, and P.-F. Rodriguez. Critical one-arm probability for the metric Gaussian free ﬁeld in low dimensions. Probab. Theory Relat. Fields , 2025

work page 2025

[11] [11]

Y. Gao, X. Li, Y. Li, R. Liu, and X. Liu. A boundary Harnack principle and its application to analyticity of 3D Brownian intersection exponents. arXiv preprint arXiv:2411.14921 , 2024

work page arXiv 2024

[12] [12]

Y. Gao, X. Li, and W. Qian. Multiple points on the boundaries of Brownian loop-soup clusters. Ann. Probab. 54(1): 216–268. , 2026. 22

work page 2026

[13] [13]

Garban, G

C. Garban, G. Pete, and O. Schramm. Pivotal, cluster, and interface measures for critical planar percolation. J. Amer. Math. Soc. , 26:939–1024, 2013

work page 2013

[14] [14]

Holden, G

N. Holden, G. Lawler, X. Li, and X. Sun. Minkowski content of Brownian cut points. Ann. Inst. H. Poincaré Probab. Statist. , 58:455–488, 2022

work page 2022

[15] [15]

S. Janson. Bounds on the distributions of extremal values of a scanning process. Stochastic Process. Appl., 18(2):313–328, 1984

work page 1984

[16] [16]

Jego and T

A. Jego and T. Lupu. Three-dimensional Brownian loop soup clusters. arXiv preprint arxiv:2601.04840, 2026

work page arXiv 2026

[17] [17]

G. Lawler. Strict concavity of the intersection exponent for Brownian motion in two and three dimensions. Math. Phys. Electron. J. , 4:paper no. 5., 1998

work page 1998

[18] [18]

Lawler, O

G. Lawler, O. Schramm, and W. Werner. Conformal restriction: the chordal case. J. Amer. Math. Soc. , 16(4):917–955, 2003

work page 2003

[19] [19]

Lawler and W

G. Lawler and W. Werner. The Brownian loop soup. Probab. Theory Relat. Fields , 128:565– 588, 2004

work page 2004

[20] [20]

G. F. Lawler. Hausdorﬀ dimension of cut points for Brownian motion. Electron. J. Probab. , 1:1–20, 1996

work page 1996

[21] [21]

G. F. Lawler. The inﬁnite two-sided loop-erased random walk. Electron. J. Probab. , 25:1–42, 2020

work page 2020

[22] [22]

G. F. Lawler, O. Schramm, and W. Werner. Values of Brownian intersection exponents. I. Half-plane exponents. Acta Math. , 187(2):237–273, 2001

work page 2001

[23] [23]

G. F. Lawler, O. Schramm, and W. Werner. Values of Brownian intersection exponents. II. Plane exponents. Acta Math. , 187(2):275–308, 2001

work page 2001

[24] [24]

G. F. Lawler, O. Schramm, and W. Werner. Analyticity of intersection exponents for planar Brownian motion. Acta Math. , 189(2):179–201, 2002

work page 2002

[25] [25]

G. F. Lawler, O. Schramm, and W. Werner. Values of Brownian intersection exponents. III. Two-sided exponents. Ann. Inst. H. Poincaré Probab. Statist. , 38(1):109–123, 2002

work page 2002

[26] [26]

G. F. Lawler and B. Vermesi. Fast convergence to an invariant measure for non-intersecting 3-dimensional Brownian paths. ALEA Lat. Am. J. Probab. Math. Stat. , 9(2):717–738, 2012

work page 2012

[27] [27]

W. Qian. Generalized disconnection exponents. Probab. Theory Relat. Fields , 179(1):117–164, 2021

work page 2021

[28] [28]

Sapozhnikov and D

A. Sapozhnikov and D. Shiraishi. On Brownian motion, simple paths, and loops. Probab. Theory Relat. Fields , 172(3):615–662, 2018

work page 2018

[29] [29]

Sheﬃeld and W

S. Sheﬃeld and W. Werner. Conformal loop ensembles: the Markovian characterization and the loop-soup construction. Annals of Mathematics , 176:1827–1917, 2012. 23

work page 1917

[30] [30]

W. Werner. On clusters of Brownian loops in d dimensions. In In and Out of Equilibrium 3: Celebrating Vladas Sidoravicius , pages 797–817. Springer, 2020

work page 2020

[31] [31]

W. Werner. A switching identity for cable-graph loop soups and Gaussian free ﬁelds. arXiv preprint arXiv:2502.06754 , 2025. 24

work page arXiv 2025