One- and two-particle spectral gap identities for the symmetric inclusion process and related models

Federico Sau; Seonwoo Kim

REVIEW 2 major objections 4 minor 1 cited by

One- and two-particle spectral gap identities for the symmetric inclusion process and related models

T0 review · 2 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read As diffusion vanishes, the k-particle inclusion process has the two-particle spectral gap; with reservoirs, the one-particle identity holds for all α_min>0.

desk verdict Sharp and mostly rigorous asymptotic two-particle reduction for SIP; the slow-fast eigenvalue step is the thing to check. read the letter →

arxiv 2412.01489 v2 pith:EVMRP6FF submitted 2024-12-02 math.PR math-phmath.MP

Seonwoo Kim , Federico Sau This is my paper · ORCID

classification math.PRmath-phmath.MP MSC 60K3560J2705C50

keywords spectralgapsymmetricinclusionprocesstwo-particleidentityslow-fastsystemsDirichletformcomparisonBrownianenergyopenrandomwalk

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

Many-particle systems with mutual attraction are hard to analyze because the particles' tendency to clump couples all degrees of freedom. This paper studies the symmetric inclusion process, a graph model in which particles jump onto occupied sites faster than onto empty ones, and asks whether the spectral gap—the rate of convergence to equilibrium—can be read off from a small subsystem. It establishes two complementary reductions: in the closed, conservative system, when the independent-diffusion parameter tends to zero, the gap of any number k≥3 of particles is asymptotically equal to the gap of two particles, while the older one-particle identity is genuinely false outside the log-concave regime α_min≥1. It also proves, by comparison with the complete graph, a sharp lower bound showing the gap decays at most linearly in the minimal site weight, rather than quadratically as the naive two-sided bound suggests. In the open system in contact with reservoirs, the gap is exactly the gap of a single random walk with killing, for every positive diffusivity, and the identities carry over to the Brownian energy process.

What carries the argument

The argument runs on two tracks. In the closed system, the engine is the rigid eigenstructure encoded by the annihilation operator a_k and its creation adjoint a†_{α,k-1}: consistency L_{G,α,k} a_k = a_k L_{G,α,k-1} lifts every eigenvalue of the (k-1)-particle system into the k-particle spectrum, and the orthogonal decomposition L²(µ_{α,k}) = Im a_k ⊕ Ker a†_{α,k-1} reduces the spectral gap to a Dirichlet-form quotient on Ker a†. The comparison Theorem 3.2 bounds the Dirichlet form of any graph by that of the complete graph using distinguishing paths and, for large stacks, two-dimensional flows; the complete-graph form satisfies the explicit identity E_{K,α,k}(f)=k(|α|+k-1)‖f‖²_{α,k}. In the vanishing-diffusivity limit the paper switches to a slow-fast decomposition $ε^{{-1}}$L_{εα̂,k}=A_{α̂,k}+$ε^{{-1}}$B_k, where B_k is the fast, purely attractive process absorbed on the set Ω_k of configurations with no two particles at distance one; projective semigroup convergence yields exact convergence of gaps to eigenvalues of the limiting generator Π_k A_{α̂,k}, whose block-triangular structure on the levels Ω_{k,m} of m separated stacks is then analyzed. A consistency relation for the limiting generators gives λ_{k,m}(α̂)=λ_{m,m}(α̂), and a variational comparison gives λ_{k,k}(α̂)≥λ_{2,2}(α̂), so the lowest transient eigenvalue is always the two-particle one. The open system instead uses orthogonal-polynomial dualities to lift eigenfunctions of the absorbing k-particle system to eigenfunctions of the reservoir process, producing a complete orthogonal basis and exact spectral gap transfer.

What would settle it

Exact diagonalization of the three- and two-particle SIP generators on a small graph, for instance the three-site path with α=(ε,ε,ε), for a sequence ε→0 should show gap_3/gap_2→1; any graph for which the numerical liminf exceeds 1 falsifies Theorem 2.3. For the complementary failure claim, on the two-dimensional torus T²_N the same kind of computation should show gap_2/gap_1→0 like 1/log N.

Watch

Extended reading notes

Core claim

The paper's central claim is that the low-lying spectrum of the conservative symmetric inclusion process is dictated by the two-particle dynamics exactly in the limit where the independent diffusion parameter vanishes. Concretely, for any finite connected graph G and fixed normalized site weights α̂, Theorem 2.3 gives lim_{ε→0} gap_k(G, ε α̂)/gap_2(G, ε α̂)=1 for every k≥3, so the many-particle spectral gap is asymptotically the two-particle gap; this is the ε-asymptotic form of the conjectured identity gap_SIP=gap_2. The companion Theorem 2.4 shows the older one-particle identity gap_SIP=gap_RW is genuinely false outside the log-concave regime α_min≥1: on large tori, the two-particle gap is strictly smaller than the random-walk gap in the vanishing-diffusivity limit. For the non-conservative process in contact with reservoirs and constant reservoir density, Corollary 6.4 restores the one-particle identity for every α_min>0: gap_SIP(G,α,ω,ρ)=gap_RW(G,α,ω), where the random walk is killed at rate ω_x at x. The same statements transfer to the Brownian energy process by the isospectrality of the two generators.

Load-bearing premise

The asymptotic two-particle identity rests on the assumption that the fast attraction mechanism always drives the system into fully piled-up configurations quickly enough, uniformly over the starting configuration, before the slow free diffusion has time to act; if a geometry allowed a long-lived intermediate configuration of separated particles, the limiting spectrum could differ.

Editorial extensions

If this is right

On every finite connected graph, the many-particle spectral gap of the closed symmetric inclusion process is, in the small-diffusivity limit, the two-particle gap; the earlier (k−1)-particle eigenvalue inclusion cannot reverse the ordering among k≥3.
The one-particle identity fails generically in the attractive regime: on large two- or higher-dimensional tori with uniform weights, gap_2/gap_1 tends to a constant below 1, with the two-particle eigenvalue of order 1/(N² log N) in two dimensions.
Even without the asymptotics, gap_SIP is controlled from below by a constant times α_min, with an explicit constant in terms of |V|, diameter, and conductivity, so the decay in diffusivity is at most linear.
Opening the system to reservoirs with a constant reservoir density restores the one-particle law: the gap equals that of a single random walk killed at the reservoir rates, for every α_min>0 and every graph.
The Brownian energy process, being isospectral with the symmetric inclusion process, inherits both the sharp lower bound and the failure or success of the one-particle identities.

Reading between the lines

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

Beyond the stated theorems, the exact identity gap_SIP=gap_2 for all α rather than only in the ε→0 limit remains open; upgrading Proposition 5.1 to a quantitative comparison with constants uniform in ε would be a natural route to prove the full conjecture for all graphs.
The dichotomy suggests a general principle for Dirichlet-reversible interacting systems: closed systems are governed by the smallest subsystem capable of forming a condensate, while reservoirs erase the interaction slowdown. This predicts the same two-particle reduction for Beta-Binomial splitting and KMP-type models on arbitrary graphs, where the paper only records partial results.
A direct testable extension: on a graph where the closed SIP gap is strictly below the random-walk gap, adding a single reservoir should replace the gap by the killed random-walk gap, discontinuously at reservoir strength zero; this could be checked numerically for moderate particle numbers.
The explicit comparison constants in Theorem 2.2 depend on |V|² diam(G) and an exponential in α_max/α_min, and the paper notes they are not optimal. On boxes and tori one expects the sharp prefactor to be governed by the ratio of relaxation time to meeting time, as in the torus calculation behind Theorem 2.4.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The short version: this is a substantial and largely rigorous paper. The genuinely new content is the asymptotic two-particle spectral gap identity in the vanishing-diffusivity limit (Theorem 2.3), the failure of the one-particle identity outside the log-concave regime on explicit geometries (Theorem 2.4), and the non-conservative one-particle identity with reservoirs (Theorem 6.1 and Corollary 6.4). The proofs are built honestly: the Dirichlet-form comparison in Section 4 is explicit, the slow-fast asymptotics in Section 5 rest on a cited limit theorem plus absorption estimates proven in Appendix A, and the paper openly says the exact two-particle identity remains unresolved. There are no fitted parameters and no circularity. The self-citation to [KS24] is appropriate; the new claims do not assume the target results.

Soft spots, in proportion: the slow-fast step is load-bearing for Theorems 2.3 and 2.4. Proposition 5.1 passes from semigroup convergence to exact convergence of the spectral gap, which requires the fast inclusion dynamics to thermalize on the absorbing set uniformly in the starting configuration. The appendix supplies explicit estimates, but they are delicate and deserve independent checking, especially the log k dependence in Lemma A.2. The overlap count in Section 4.5 is another place where the text says "tedious computations" and waves at Case 3; it is probably correct, but a referee should verify it. The non-reversible extension in Remark 6.5 is flagged by the authors as plausible but not a theorem, which is acceptable but limits the scope of Corollary 6.4. None of these issues undermines the main claims as far as I can tell, but Section 4 and Appendix A are where a referee should spend time.

Who this is for: researchers working on spectral gaps, metastability, and interacting particle systems, particularly those interested in Caputo's conjecture and its variants. It adds SIP to the short list of models where a spectral gap identity can be proved or sharply disproved on general geometries. I would send it to a serious referee and, conditional on a careful check of Sections 4 and 5, accept it. My own verdict is close to accept, with the main caveat being independent verification of the technical estimates.

Referee Report

2 major / 4 minor

Summary. The paper studies the spectral gap of the symmetric inclusion process (SIP) on finite weighted graphs, in both conservative and open (reservoir) settings, and transfers some results to the Brownian energy process. The main results are: (i) Theorem 2.2, a non-asymptotic lower bound for gap_{SIP}(G,α) that is linear in α_min and thus sharp up to constants depending only on the geometry; (ii) Theorem 2.3, an asymptotic two-particle spectral gap identity lim_{ε→0} gap_k(G,ε α̂)/gap_2(G,ε α̂) = 1 for every k ≥ 3; (iii) Theorem 2.4, giving graphs for which the one-particle identity fails, with lim gap_2/gap_1 < 1; and (iv) Theorem 6.1 and Corollary 6.4, a one-particle spectral gap identity for the non-conservative SIP with absorbing or reversible reservoirs, valid for all α_min > 0. The proofs combine the rigid eigenstructure and annihilation/creation operators of SIP, Dirichlet-form comparison inequalities via constructed paths and flows, slow-fast limit theorems with absorption estimates, and orthogonal-polynomial duality for the reservoir system.

Significance. If the proofs are correct, these are substantial contributions to the spectral theory of interacting particle systems. The two-particle asymptotic identity is a new structural result that gives a precise sense in which the two-particle dynamics dominates the relaxation of the many-particle conservative SIP in the low-diffusivity limit; the sharp α_min-linear lower bound resolves a natural scaling question left open by the earlier quadratic bound (1.5). The non-conservative one-particle identity extends the short list of models satisfying Aldous-type spectral gap identities to a strongly interacting, genuinely reversible open system. The paper is largely self-contained: the main arguments are explicit, the comparison constants are explicit, and the slow-fast analysis is supported by two technical appendices. The paper does not rely on fitted parameters or circular assumptions. The main weaknesses are localized technical points in the proof of Proposition A.1 and in the overlap-counting step of Section 4.5, both of which appear repairable.

major comments (2)

[Appendix A, Lemma A.4] The proof asserts that τ_{∆_k}, the first hitting time of the transient set, stochastically dominates an exponential random variable with rate c_max k α̂_max. This is not correct as stated: from a configuration ξ ∈ Ω_k, the total rate of jumps that enter ∆_k is generally larger than c_max k α̂_max by a factor comparable to the maximum degree of G, because several directed edges can contribute to entering ∆_k simultaneously. Consequently the displayed stochastic domination fails. The intended conclusion, however, follows from the same argument after replacing the single-exponential domination by a union bound using a uniform upper bound on the total jump rate into ∆_k, so Proposition A.1 should remain valid once the lemma is corrected. Since Proposition A.1 is used in Proposition 5.1 and hence in Corollary 5.2, this correction is load-bearing.
[Section 4.5, Case 3] The proof of (4.3) relies on the claim that each Dirichlet summand ∇²f(η, η − δ_z + δ_w) is produced by at most six triples (ℓ, m, σ), and Case 3, the reverse orientation (x_{s−1}, x_s) = (w, z), is dismissed as 'almost identical' without details. Because the constant 6 enters the final comparison constant of Theorem 3.2, a full verification of Case 3, or a more systematic counting argument covering both orientations, should be supplied so that the reader can check the bound.

minor comments (4)

[Section 7.2] There are typographical errors in the derivation after (7.6): 'ω_x ξ_x (θ_x − σ)' should read 'ω_x ξ_x (θ_x − ̺)', and 'D_σ(ξ − δ_x, η)' should read 'D_{α,̺}(ξ − δ_x, η)'; later in the same display, 'F^{b̺_{α,ω,θ,k−1}}_{α,̺}' should be 'F^{b^θ_{α,ω,̺,k−1}}_{α,̺}'.
[Section 6.2] The proof of Theorem 6.1 establishes gap_k(G,α,ω) ≥ gap_1(G,α,ω) via the survival-probability bound (7.2), but the reverse inequality gap_k(G,α,ω) ≤ gap_1(G,α,ω), which is needed for equality, is not explicitly proved; it follows from the annihilation/lifting intertwining for the killed dynamics, but should be stated.
[Section 5.5] In the proof of Theorem 2.4, the inequalities (5.29) and the use of [Her23, Eq. (1.7)] are quoted from external sources without displaying the exact statement used; please state the precise inequalities so that the reader can verify the direction of the bound λ_{2,2} ≤ 1/E_{κ(·|W)}[τ_{W^c}].
[Section 4.5] The paragraph after (Case 2.III) says that forward jumps in type [III] are of three types, but the correspondence between these three types and the arrows in Figure 4.3 is not fully spelled out; a short table or explicit map from the figure's arrows to the three cases would improve reproducibility of the count.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the main spectral-gap results are derived from independent slow-fast limits, Dirichlet-form comparisons, and algebraic consistency identities, with self-citations used only as prior published tools.

full rationale

The paper's central claims do not reduce to their inputs. Theorem 2.2 follows from a complete-graph spectral computation (Proposition 3.1), a graph-to-complete-graph Dirichlet-form comparison (Theorem 3.2), and the orthogonal decomposition lifted from [KS24]; none of these assumes the target lower bound. Theorems 2.3 and 2.4 are proved through the slow-fast semigroup convergence of Proposition 5.1, imported from [Kur73]/[EK86] rather than from the paper's conclusions, followed by independent spectral analyses of the limiting metastable chain: block-triangular decomposition, consistency inherited from SIP, and a variational comparison showing lambda_{k,k} >= lambda_{2,2}. The non-conservative identity in Theorem 6.1 is proved by a lookdown representation and survival-probability domination, not by assuming the gap identity; Corollary 6.4 then follows from the orthogonal polynomial basis in Theorem 6.3. The self-citations to [KS24] provide the nested eigenstructure and consistency identities, but these are prior published, checkable results used as lemmas, not as substitutes for the new conclusions. There are no fitted parameters, no quantity called a prediction that is actually a fit, and no uniqueness theorem imported from the authors' own earlier work to force a choice. The admitted open point in Remark 6.5 concerns the non-reversible non-conservative setting and is a limitation, not a circular step.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

No fitted parameters or invented entities. The results rest on standard Markov chain theory, slow-fast limit theorems, and prior structural results from [KS24] and [FRS22]; none of these presuppose the target theorems.

assumptions (5)

standard math SIP_k(G,alpha) is reversible with respect to the Dirichlet-Multinomial measure (2.2), and its spectral gap admits the variational characterization (3.4).
Used in Section 3.1 for the eigenstructure and gap formulas; derived by a standard detailed balance computation given in Section 2.1.
domain assumption The consistency identity L_{G,alpha,k} a_k = a_k L_{G,alpha,k-1} and the adjoint creation operator a†_{alpha,k-1} with orthogonal decomposition (3.3) hold; these are imported from [KS24].
This is the main self-cited prior input. It grounds the spectral interleaving (2.6), the reduction (3.5), and the complete-graph normalization in Proposition 3.1.
standard math The slow-fast limit theorem of Kurtz and Ethier-Kurtz applies to epsilon^{-1}L_{epsilon alpha_hat,k} = A_{alpha_hat,k} + epsilon^{-1}B_k (Eq. 5.1), yielding semigroup convergence (5.4) and eigenvalue asymptotics (Corollary 5.2).
Invoked in Section 5.1. The absorption estimates in Appendix A are proved to ensure the projection term vanishes.
domain assumption The orthogonal polynomial duality functions D_{alpha,rho}(xi,eta) from [FRS22] form a complete orthogonal system in L2(nu_{alpha,rho}) and satisfy the duality relation (7.6).
Used in Section 7.2 to prove Theorem 6.3 and Corollary 6.4; [FRS22] is prior literature with overlapping authorship.
standard math Known random walk estimates on the discrete torus: relaxation time bound (5.29) and meeting time asymptotics from [Cox89, LP17, Her23].
Used in the proof of Theorem 2.4 to separate gap_RW from lambda_{2,2}.

how reviews work

0 comments

Cite this review

Pith. "Pith review of One- and two-particle spectral gap identities for the symmetric inclusion process and related models." pith.science (2026). https://pith.science/paper/EVMRP6FF

@misc{pith2026241201489,
  author       = {Pith},
  title        = {Pith review of: One- and two-particle spectral gap identities for the symmetric inclusion process and related models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EVMRP6FF}},
  note         = {Machine review of arXiv:2412.01489}
}

read the original abstract

The symmetric inclusion process (SIP) models particles diffusing on a graph with mutual attraction. We recently showed that, in the log-concave regime (where diffusivity dominates interaction), the spectral gap of the conservative SIP matches that of a single particle. In this paper, our main result demonstrates that this identity generally fails outside this regime, but always holds for the non-conservative SIP, regardless of the interaction strength. When this one-particle spectral gap identity breaks down, we derive sharp bounds for the gap in terms of diffusivity, and reveal a two-particle spectral gap identity in the vanishing diffusivity limit. Our approach leverages the rigid eigenstructure of SIP, refined comparisons of Dirichlet forms for arbitrary diffusivity and particle numbers, and techniques from slow-fast system analysis. These findings extend to the dual interacting diffusion known as Brownian energy process, and shed some light on the spectral gap behavior for related Dirichlet-reversible systems on general, non-mean-field, geometries.

Figures

Figures reproduced from arXiv: 2412.01489 by the authors.

**Figure 4.1.** Path of configurations from σ xy m,ℓ−m to σ xy m−1,ℓ−m+1 in Section 4.2. In this occupied case, the red particle simply moves from x to y along the sequence x = x0, x1, . . . , xt = y consecutively. 4.2. Decomposition of gradients: occupied sites. Now, assume that (cf. (4.10)) cxy = 0 , such that t ≥ 2 . (4.11) Additionally, in this subsection, assume that all sites along the sequence (4.7) are occupied by particles… view at source ↗

**Figure 4.2.** Path of configurations from σ xy m,ℓ−m to σ xy m−1,ℓ−m+1 in Section 4.3. Double arrows indicate series of consecutive jumps. Here, first the stack of m particles at site x moves to xt−1 (step (F)), then the red particle jumps from site xt−1 to y (step (S)), and then the remaining stack of m−1 particles at site xt−1 moves back to x (step (B)). path from σ xy m,ℓ−m to σ xy m−1,ℓ−m+1. Applying (4.14) along the path con… view at source ↗

**Figure 4.3.** Two-dimensional system of paths from σ xy m,ℓ−m (top left) to σ xy m−1,ℓ−m+1 (top right) explained in Section 4.4. Double arrows indicate series of consecutive paths. Moreover, the blue (resp. red) numbers indicate the corresponding value of the flow ϕ along the path in the vertical (resp. horizontal) direction, as defined in (4.31) and (4.32). The values are chosen so as to ensure that ϕ is a unit flow from σ xy m,… view at source ↗

Figures from the paper (3 more)

**Figure 4.4.** Figure 4.4: Path from σ xy m,ℓ−m to σ xy m−1,ℓ−m+1 in the general case explained in Section 4.5. The figure explains how the red particle moves from x0 = x to xt = y along the path. First, in O1 where the sites are occupied, the red particle simply jumps consecutively to the rig…

**Figure 5.1.** Figure 5.1: The H-shape graph G and the corresponding absorbing spaces Ω4,4 = Ω1 4,4 and Ω5,4 = Ω1,1 5,4 ⊔ Ω 1,2 5,4 explained in Remark 5.10. Remark 5.10. The irreducible structure in (5.18) may indeed get strictly finer than the one in (5.16). For example, consider the followi…

**Figure 5.2.** Figure 5.2: In this figure, we depict the eigenvalue equation in (5.23) (for simplicity, we dropped ˆα from the notation). On the left-hand side, we show the |Ωk| × |Ωk| matrix Mα,k ˆ in its block lower triangular form. Stars (⋆) indicate some possibly non-zero entries; triangle…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

Spectral gap of the KMP and other stochastic exchange models on arbitrary graphs
math.PR 2025-05 conditional novelty 8.0 of 10

The spectral gap of the KMP, harmonic, and immediate exchange models on arbitrary graphs is comparable to, and sometimes exactly equal to, the spectral gap of a natural random walk.

Reference graph

Works this paper leans on

53 extracted references · 50 canonical work pages · cited by 1 Pith paper

[1]

Alcaraz, Michel Droz, Malte Henkel, and Vladimir Rittenberg

Francisco C. Alcaraz, Michel Droz, Malte Henkel, and Vladimir Rittenberg. Reaction-diffusion processes, critical dynamics, and quantum chains. Ann. Physics , 230(2):250--302, 1994

work page 1994
[2]

R eversible M arkov C hains and R andom W alks on G raphs, 2002

David Aldous and James Allen Fill. R eversible M arkov C hains and R andom W alks on G raphs, 2002. Unfinished monograph, recompiled 2014, available at https://www.stat.berkeley.edu/users/aldous/RWG/book.pdf

work page 2002
[3]

Entropy inequalities for random walks and permutations

Alexandre Bristiel and Pietro Caputo. Entropy inequalities for random walks and permutations. Ann. Inst. Henri Poincar\' e Probab. Stat. , 60(1):54--81, 2024

work page 2024
[4]

Metastability in the reversible inclusion process

Alessandra Bianchi, Sander Dommers, and Cristian Giardin\`a. Metastability in the reversible inclusion process. Electron. J. Probab. , 22:Paper No. 70, 34, 2017

work page 2017
[5]

Metastability , volume 351 of Grundlehren der mathematischen Wissenschaften

Anton Bovier and Frank den Hollander. Metastability , volume 351 of Grundlehren der mathematischen Wissenschaften . Springer, Cham, 2015. A potential-theoretic approach

work page 2015
[6]

On the spectral gap of the K ac walk and other binary collision processes

Pietro Caputo. On the spectral gap of the K ac walk and other binary collision processes. ALEA Lat. Am. J. Probab. Math. Stat. , 4:205--222, 2008

work page 2008
[7]

E. A. Carlen, M. C. Carvalho, and M. Loss. Determination of the spectral gap for K ac's master equation and related stochastic evolution. Acta Math. , 191(1):1--54, 2003

work page 2003
[8]

Duality for stochastic models of transport

Gioia Carinci, Cristian Giardin\`a, Claudio Giberti, and Frank Redig. Duality for stochastic models of transport. J. Stat. Phys. , 152(4):657--697, 2013

work page 2013

Show all 53 references

[9]

Mixing time of the adjacent walk on the simplex

Pietro Caputo, Cyril Labb\' e , and Hubert Lacoin. Mixing time of the adjacent walk on the simplex. Ann. Probab. , 48(5):2449--2493, 2020

2020
[10]

Liggett, and Thomas Richthammer

Pietro Caputo, Thomas M. Liggett, and Thomas Richthammer. Proof of A ldous' spectral gap conjecture. J. Amer. Math. Soc. , 23(3):831--851, 2010

2010
[11]

On the spectrum of a neutral multi-allelic M oran model

Josu \'e Corujo. On the spectrum of a neutral multi-allelic M oran model. ALEA Lat. Am. J. Probab. Math. Stat. , 20(1):505--546, 2023

2023
[12]

J. T. Cox. Coalescing random walks and voter model consensus times on the torus in Z ^d . Ann. Probab. , 17(4):1333--1366, 1989

1989
[13]

Derrida, M

B. Derrida, M. R. Evans, V. Hakim, and V. Pasquier. Exact solution of a 1 D asymmetric exclusion model using a matrix formulation. J. Phys. A , 26(7):1493--1517, 1993

1993
[14]

Ferrari, and Davide Gabrielli

Anna De Masi, Pablo A. Ferrari, and Davide Gabrielli. Hidden T emperature in the KMP M odel. J. Stat. Phys. , 191(11):Paper No. 150, 2024

2024
[15]

Mathematical methods for hydrodynamic limits , volume 1501 of Lecture Notes in Mathematics

Anna De Masi and Errico Presutti. Mathematical methods for hydrodynamic limits , volume 1501 of Lecture Notes in Mathematics . Springer-Verlag, Berlin, 1991

1991
[16]

Comparison theorems for reversible M arkov chains

Persi Diaconis and Laurent Saloff-Coste. Comparison theorems for reversible M arkov chains. Ann. Appl. Probab. , 3(3):696--730, 1993

1993
[17]

D ynamics on Graphs , 2024

Rick Durrett. D ynamics on Graphs , 2024. unpublished monograph, version of the 16th of A pril 2024, available at https://services.math.duke.edu/ rtd/DoG/DoG.html

2024
[18]

Ethier and Thomas G

Stewart N. Ethier and Thomas G. Kurtz. Markov processes. Characterization and convergence . Wiley Series in Probability and Mathematical Statistics: Probability and Mathematical Statistics. John Wiley & Sons, Inc., New York, 1986

1986
[19]

Integrable heat conduction model

Chiara Franceschini, Rouven Frassek, and Cristian Giardin\`a. Integrable heat conduction model. J. Math. Phys. , 64(4):Paper No. 043304, 21, 2023

2023
[20]

Duality and hidden equilibrium in transport models

Rouven Frassek, Cristian Giardin\`a, and Jorge Kurchan. Duality and hidden equilibrium in transport models. SciPost Phys. , 9(4):Paper No. 054, 18, 2020

2020
[21]

Symmetric inclusion process with slow boundary: hydrodynamics and hydrostatics

Chiara Franceschini, Patr\' cia Gon c alves, and Federico Sau. Symmetric inclusion process with slow boundary: hydrodynamics and hydrostatics. Bernoulli , 28(2):1340--1381, 2022

2022
[22]

Orthogonal polynomial duality of boundary driven particle systems and non-equilibrium correlations

Simone Floreani, Frank Redig, and Federico Sau. Orthogonal polynomial duality of boundary driven particle systems and non-equilibrium correlations. Ann. Inst. Henri Poincar\' e Probab. Stat. , 58(1):220--247, 2022

2022
[23]

Duality and exact correlations for a model of heat conduction

Cristian Giardin\`a, Jorge Kurchan, and Frank Redig. Duality and exact correlations for a model of heat conduction. J. Math. Phys. , 48(3):033301, 15, 2007

2007
[24]

Duality and hidden symmetries in interacting particle systems

Cristian Giardin\`a, Jorge Kurchan, Frank Redig, and Kiamars Vafayi. Duality and hidden symmetries in interacting particle systems. J. Stat. Phys. , 135(1):25--55, 2009

2009
[25]

Intertwining and propagation of mixtures for generalized KMP models and harmonic models

Cristian Giardin \`a , Frank Redig, and Berend van Tol. Intertwining and propagation of mixtures for generalized KMP models and harmonic models. arXiv:2406.01160 , 2024

2024 arXiv
[26]

Dynamics of condensation in the symmetric inclusion process

Stefan Grosskinsky, Frank Redig, and Kiamars Vafayi. Dynamics of condensation in the symmetric inclusion process. Electron. J. Probab. , 18:no. 66, 23, 2013

2013
[27]

Relaxation times are stationary hitting times of large sets

Jonathan Hermon. Relaxation times are stationary hitting times of large sets. arXiv:2304.05878 , 2023

2023 arXiv
[28]

M. Kac. Foundations of kinetic theory. In Proceedings of the T hird B erkeley S ymposium on M athematical S tatistics and P robability, 1954--1955, vol. III , pages 171--197. Univ. California Press, Berkeley-Los Angeles, Calif., 1956

1954
[29]

Second time scale of the metastability of reversible inclusion processes

Seonwoo Kim. Second time scale of the metastability of reversible inclusion processes. Probab. Theory Related Fields , 180(3-4):1135--1187, 2021

2021
[30]

Hierarchical structure of metastability in the reversible inclusion process: third time scale and complete characterization

Seonwoo Kim. Hierarchical structure of metastability in the reversible inclusion process: third time scale and complete characterization. arXiv:2308.13842 , 2023

2023 arXiv
[31]

Scaling limits of interacting particle systems , volume 320 of Grundlehren der Mathematischen Wissenschaften

Claude Kipnis and Claudio Landim. Scaling limits of interacting particle systems , volume 320 of Grundlehren der Mathematischen Wissenschaften . Springer-Verlag, Berlin, 1999

1999
[32]

Lesky, and Ren\'e F

Roelof Koekoek, Peter A. Lesky, and Ren\'e F. Swarttouw. Hypergeometric orthogonal polynomials and their q -analogues . Springer Monographs in Mathematics. Springer-Verlag, Berlin, 2010. With a foreword by Tom H. Koornwinder

2010
[33]

Kipnis, C

C. Kipnis, C. Marchioro, and E. Presutti. Heat flow in an exactly solvable model. J. Statist. Phys. , 27(1):65--74, 1982

1982
[34]

Condensation and metastable behavior of non-reversible inclusion processes

Seonwoo Kim and Insuk Seo. Condensation and metastable behavior of non-reversible inclusion processes. Comm. Math. Phys. , 382(2):1343--1401, 2021

2021
[35]

Spectral gap of the symmetric inclusion process

Seonwoo Kim and Federico Sau. Spectral gap of the symmetric inclusion process. Ann. Appl. Probab. , 34(5):4899--4920, 2024

2024
[36]

Thomas G. Kurtz. A limit theorem for perturbed operator semigroups with applications to random evolutions. J. Functional Analysis , 12:55--67, 1973

1973
[37]

Metastable M arkov chains: from the convergence of the trace to the convergence of the finite-dimensional distributions

Claudio Landim, Michail Loulakis, and Mustapha Mourragui. Metastable M arkov chains: from the convergence of the trace to the convergence of the finite-dimensional distributions. Electron. J. Probab. , 23:Paper No. 95, 34, 2018

2018
[38]

Levin and Yuval Peres

David A. Levin and Yuval Peres. Markov chains and mixing times . American Mathematical Society, Providence, RI, 2017. Second edition of [MR2466937], with contributions by Elizabeth L. Wilmer, with a chapter on ``Coupling from the past'' by James G. Propp and David B. Wilson

2017
[39]

Mathematical aspects of mixing times in M arkov chains

Ravi Montenegro and Prasad Tetali. Mathematical aspects of mixing times in M arkov chains. Found. Trends Theor. Comput. Sci. , 1(3):x+121, 2006

2006
[40]

Mixing of the symmetric beta-binomial splitting process on arbitrary graphs

Richard Pymar and Nicol \'a s Rivera. Mixing of the symmetric beta-binomial splitting process on arbitrary graphs. arXiv:2307.02406 , 2023

2023 arXiv
[41]

Mixing of the averaging process and its discrete dual on finite-dimensional geometries

Matteo Quattropani and Federico Sau. Mixing of the averaging process and its discrete dual on finite-dimensional geometries. Ann. Appl. Probab. , 33(2):936--971, 2023

2023
[42]

Factorized duality, stationary product measures and generating functions

Frank Redig and Federico Sau. Factorized duality, stationary product measures and generating functions. J. Stat. Phys. , 172(4):980--1008, 2018

2018
[43]

Universality of cutoff for exclusion with reservoirs

Justin Salez. Universality of cutoff for exclusion with reservoirs. Ann. Probab. , 51(2):478--494, 2023

2023
[44]

Spectral gap and curvature of monotone M arkov chains

Justin Salez. Spectral gap and curvature of monotone M arkov chains. Ann. Probab. , 52(3):1153--1161, 2024

2024
[45]

Spectral gap for stochastic energy exchange model with nonuniformly positive rate function

Makiko Sasada. Spectral gap for stochastic energy exchange model with nonuniformly positive rate function. Ann. Probab. , 43(4):1663--1711, 2015

2015
[46]

Lectures on finite M arkov chains

Laurent Saloff-Coste. Lectures on finite M arkov chains. In Lectures on probability theory and statistics ( S aint- F lour, 1996) , volume 1665 of Lecture Notes in Math. , pages 301--413. Springer, Berlin, 1997

1996
[47]

G. M. Sch\" u tz. Exactly solvable models for many-body systems far from equilibrium. In Phase transitions and critical phenomena, V ol. 19 , pages 1--251. Academic Press, San Diego, CA, 2001

2001
[48]

\' E quations diff\' e rentielles provenant de la g\' e n\' e tique des populations

Norio Shimakura. \' E quations diff\' e rentielles provenant de la g\' e n\' e tique des populations. T\^ o hoku Math. J. , 29(2):287--318, 1977

1977
[49]

A G ibbs sampler on the n -simplex

Aaron Smith. A G ibbs sampler on the n -simplex. Ann. Appl. Probab. , 24(1):114--130, 2014

2014
[50]

Interaction of M arkov processes

Frank Spitzer. Interaction of M arkov processes. Advances in Math. , 5:246--290 (1970), 1970

1970
[51]

Long range correlations for stochastic lattice gases in a nonequilibrium steady state

Herbert Spohn. Long range correlations for stochastic lattice gases in a nonequilibrium steady state. J. Phys. A , 16(18):4275--4291, 1983

1983
[52]

Large scale dynamics of interacting particles

Herbert Spohn. Large scale dynamics of interacting particles . Theoretical and Mathematical Physics. Springer Berlin, Heidelberg, 1991

1991
[53]

Nash inequality for diffusion processes associated with D irichlet distributions

Feng-Yu Wang and Weiwei Zhang. Nash inequality for diffusion processes associated with D irichlet distributions. Front. Math. China , 14(6):1317--1338, 2019

2019

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

Alcaraz, Michel Droz, Malte Henkel, and Vladimir Rittenberg

Francisco C. Alcaraz, Michel Droz, Malte Henkel, and Vladimir Rittenberg. Reaction-diffusion processes, critical dynamics, and quantum chains. Ann. Physics , 230(2):250--302, 1994

work page 1994

[2] [2]

R eversible M arkov C hains and R andom W alks on G raphs, 2002

David Aldous and James Allen Fill. R eversible M arkov C hains and R andom W alks on G raphs, 2002. Unfinished monograph, recompiled 2014, available at https://www.stat.berkeley.edu/users/aldous/RWG/book.pdf

work page 2002

[3] [3]

Entropy inequalities for random walks and permutations

Alexandre Bristiel and Pietro Caputo. Entropy inequalities for random walks and permutations. Ann. Inst. Henri Poincar\' e Probab. Stat. , 60(1):54--81, 2024

work page 2024

[4] [4]

Metastability in the reversible inclusion process

Alessandra Bianchi, Sander Dommers, and Cristian Giardin\`a. Metastability in the reversible inclusion process. Electron. J. Probab. , 22:Paper No. 70, 34, 2017

work page 2017

[5] [5]

Metastability , volume 351 of Grundlehren der mathematischen Wissenschaften

Anton Bovier and Frank den Hollander. Metastability , volume 351 of Grundlehren der mathematischen Wissenschaften . Springer, Cham, 2015. A potential-theoretic approach

work page 2015

[6] [6]

On the spectral gap of the K ac walk and other binary collision processes

Pietro Caputo. On the spectral gap of the K ac walk and other binary collision processes. ALEA Lat. Am. J. Probab. Math. Stat. , 4:205--222, 2008

work page 2008

[7] [7]

E. A. Carlen, M. C. Carvalho, and M. Loss. Determination of the spectral gap for K ac's master equation and related stochastic evolution. Acta Math. , 191(1):1--54, 2003

work page 2003

[8] [8]

Duality for stochastic models of transport

Gioia Carinci, Cristian Giardin\`a, Claudio Giberti, and Frank Redig. Duality for stochastic models of transport. J. Stat. Phys. , 152(4):657--697, 2013

work page 2013