Pith. sign in

REVIEW 4 major objections 5 minor 39 references

Metastable Hierarchy in Abstract Low-Temperature Lattice Models

T0 review · 4 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read For any abstract lattice system with Metropolis-type dynamics, the stable plateaus organize into a finite hierarchy, and at each level the accelerated tunneling dynamics converges to a simple Markov chain.

desk verdict A clean, useful survey of metastable hierarchy in Metropolis lattice systems, but the main theorem and its key separation lemma are unproved here and relegated to the author's own preprint. read the letter →

arxiv 2502.00073 v1 pith:J7GSKAIN submitted 2025-01-31 math.PR

Seonwoo Kim This is my paper · ORCID
classification math.PR MSC 60K3582C2282C20
keywords metastabilitylow-temperaturelatticemodelshierarchicaldecompositionstableplateauxenergybarriersMetropolis-typedynamicsIsingmodeltraceMarkovchain
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

This paper claims that metastability in low-temperature lattice systems is layered rather than a single rare event. For any finite lattice configuration space with a Hamiltonian and Metropolis-type jump rates, the stable plateaus can be grouped recursively into cycles according to the smallest energy barrier separating them. Each level $h$ has its own time scale $e^{\Gamma^{\star,h}\beta}$; after accelerating by that factor, the tunneling between the deepest valley groups converges to a simple finite Markov chain, while the original process spends negligible time outside those valleys. The number of recurrent groups strictly decreases level by level, so the procedure terminates at a unique class that consists exactly of the ground states. If this is right, a complicated high-dimensional energy landscape is reduced to a finite stack of simple tunneling processes.

What carries the argument

The load-bearing mechanism is an iterative cycle contraction. A stable plateau is a connected set of equal-energy configurations whose boundary has strictly higher energy; the communication height $\Phi(\eta,\xi)$ is the minimum over paths of the maximum energy along the path. At each level $h$, the plateaux are grouped into disjoint cycles $V_i^h$ whose depth equals the current minimal barrier $\Gamma^{\star,h}$, and the contracted graph and trace-chain construction produces a Markov chain on the recurrent groups at that level. The argument depends on the strict monotonicity $\Gamma^{\star,h}>\Gamma^{\star,h-1}$, which keeps the exponential time scales $e^{\Gamma^{\star,h}\beta}$ separated; deeper plateaux act as absorbing states while the trace process encodes tunneling between plateaux of depth at least $\Gamma^{\star,h}$.

What would settle it

Compute the first two minimal energy barriers $\Gamma^{\star,1}$ and $\Gamma^{\star,2}$ in any finite Metropolis lattice model; equality would make the two exponential time scales coincide, so the accelerated processes could not converge to distinct level-1 and level-2 limit chains, contradicting the claimed hierarchy.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that every abstract Metropolis-type lattice model carries a canonical hierarchical decomposition of metastability. Starting from the stable plateaux $P^1$, define the depth of a plateau as the smallest energy barrier that must be crossed to reach any other plateau; plateaux of minimal depth form the first recurrent groups, and the trace of the accelerated dynamics on them converges to a limit chain $X^{\star,1}$. Repeating the construction on the irreducible components produces levels $h=1,\dots,\mathfrak{m}$ with strictly increasing minimal depths $\Gamma^{\star,h}$; Theorem 3 asserts that for each $h$ the $e^{\Gamma^{\star,h}\beta}$-accelerated trace process on the recurrent plateau groups converges to the finite Markov chain $X^{\star,h}$, and the original process spends negligible time outside the union of the level-$h$ valleys. The terminal level's unique recurrent class is exactly the set of ground states. The second half of the paper checks that this hierarchy is realized in four Ising-model settings and identifies in each case the explicit time scales and plateau classes.

Load-bearing premise

The level-by-level hierarchy depends on the unproved assertion that the minimal energy barrier at each level is strictly larger than the one before it, since equal barriers would merge the exponential time scales and collapse the separate Markov-chain limits.

Editorial extensions

If this is right

  • If the central theorem is correct, metastable behavior in any Metropolis lattice model is described by a finite list of energy barriers and a limit Markov chain at each level.
  • The limiting chain at each level is finite-state and built from the trace process, so asymptotic hitting probabilities and transition rates become finite-dimensional, computable objects.
  • The ground states are recurrent at every level, and the unique recurrent class of the terminal level consists exactly of the ground states.
  • In the four Ising examples, the levels correspond to distinct physical mechanisms: edge flips and droplet shrinkage, droplet growth, strip formation, and the final transition to the ground state.
  • The construction uses only the Hamiltonian and the Metropolis rates, so the paper suggests the same hierarchy should transfer to broader rare-transition dynamics beyond the reversible Metropolis class.

Reading between the lines

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

  • A practical consequence that the paper leaves implicit is that the hierarchy could be computed algorithmically from finite portions of the energy landscape, turning metastability classification into a combinatorial search over droplet geometries.
  • If a model were found with equal consecutive minimal depths, the levels would merge rather than disappear, so the hierarchy would survive in coarsened form rather than being refuted outright.
  • The same recursive grouping should apply to any reversible finite Markov chain whose transition rates are exponentially small in a parameter, with stable plateaus replaced by the bottom sets of cycles.
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.

Referee Report

4 major / 5 minor

Summary. The paper announces a general theorem (Theorem 3) asserting that for any finite-state Metropolis lattice model, the collection of stable plateaux admits a hierarchical decomposition into finitely many levels, with the accelerated trace process converging to a simple Markov chain at each level. It presents the formalism of stable plateaux, cycles, and contracted/trace Markov chains in Section 2, and then applies this framework to four Ising-model settings: Glauber dynamics with positive and zero external fields and Kawasaki dynamics with few and many particles, stated as Theorems 4–11. The paper explicitly states that most proofs are omitted and refers to the author's earlier preprint [22] for the hierarchical decomposition and to other works for the concrete cases.

Significance. If Theorem 3 and its supporting lemmas are correct, the proposed framework would provide a useful unifying perspective on metastable tunneling in low-temperature lattice systems. The four examples cover classical settings, and several of the example results are drawn from published works, which lends some plausibility. However, the central general theorem and the key structural lemmas are stated without proof, and the proof is deferred to the author's own unpublished preprint [22]. The manuscript does not contain machine-checked proofs, reproducible code, or an independent verification of the main claim. Therefore, the significance of the paper as a contribution to the mathematical literature cannot be assessed from this document alone; it currently functions as an announcement of results proved elsewhere.

major comments (4)
  1. [Section 2.1, Lemma 4] Lemma 4, stated immediately after equation (28), asserts the strict separation of time scales, Γ*_h > Γ*_{h-1}, but no proof is provided. This lemma is load-bearing: if two consecutive minimal depths coincide, the exponential time scales merge, the trace process at level h is not separated from level h−1, and the claimed simple Markov chain limits at each level lose their meaning. The text defers to [22], an unpublished preprint by the author. The paper should either include a proof or a proof sketch of Lemma 4, or clearly identify this statement as a conjecture or as a result announced in [22] so that the reader can judge its status.
  2. [Section 2.2, Theorem 3] In the statement of Theorem 3, the notion of convergence is not specified. The phrase 'converges' should be made precise, for example as weak convergence of the finite-dimensional distributions of the accelerated process in the Skorokhod space D([0,T], P^{*,h}) with the J1 topology, or as convergence of the full process in a suitable path space. The second assertion about the negligible time outside V^{*,h} also needs a precise interpretation of the expectation over the initial state. Without a precise statement, the theorem cannot be checked or applied.
  3. [Section 2.1, Theorems 1–2 and Lemmas 3, 5, 6] The inductive construction of the hierarchy depends on several statements that are asserted without proof: Lemma 3 (equal energy inside P_i^h), Lemma 5 (disjointness and cycle structure of V_i^h), Lemma 6 (identification of (C^h)^⋆ and (C^h)^♯), Theorem 1 (strict decrease of the number of recurrent components), and Theorem 2 (ground-state recurrence). These are not mere technicalities: Theorem 1 guarantees that the procedure terminates, and Theorem 2 identifies the terminal level. Since these results are central to the claimed hierarchy, the manuscript should at least provide proof sketches or exact references to proofs in [22] for each statement.
  4. [Section 3, Theorems 9–11] The fourth example (Kawasaki dynamics with many particles) is stated as a theorem, but its content, including the definitions of the stable plateaux in Theorem 9 and the hierarchical decomposition in Theorem 10, is deferred entirely to the author's preprint [22]. As a result, the reader cannot verify this application from the present manuscript. The paper should clearly distinguish between results that are established in the published literature and results that are announced from [22], and should provide enough detail to make the statements self-contained or point to a verifiable source.
minor comments (5)
  1. [Section 1] The introduction refers to 'Section ??, where we discuss some possible extensions', but no such section appears in the manuscript; this placeholder should be resolved.
  2. [Section 2, Definition 4] In equation (15), the rate R_C(F(C), η) is defined using |F(C)|^{-1}; the paper should explicitly note that this corresponds to choosing an element of F(C) uniformly at random when leaving the contracted cycle.
  3. [Section 3.1] The notation P^{*,k} in Section 3.1 denotes a set of configurations, while in Section 2 the same symbol denotes a collection of plateau sets. This conflation should be clarified, for example by stating that in the positive-field example every stable plateau is a singleton so that the collection can be identified with its union.
  4. [Equations (30) and Lemma 6] The condition expressing that a previous cycle does not intersect any new cycle is written with a symbol that may be misprinted as '= / 0'; the intended condition 'C ∩ V_i^h = ∅' should be stated unambiguously.
  5. [Throughout] Several typographical errors are present, such as 't-em perature' in the abstract; the manuscript would benefit from a careful proofreading pass.

Circularity Check

2 steps flagged · score 6.0 of 10

Central hierarchy theorem and its strict time-scale separation lemma are unproved and deferred to the author's own preprint [22], making the universal claim dependent on a self-citation chain.

  1. self citation load bearing [Section 2.1 (after Eq. (28), Lemma 4) and Section 2.2 (Theorem 3); Introduction p.3]
    "The results are based on the works from [34, 3] (first), [33, 24] (second), [7] (third), and [22] (fourth). In this review article, we omit most of the proofs of the main results; interested readers are referred to the full version [22] for the hierarchical decomposition and the above-mentioned references for the concrete examples. ... Lemma 4. It holds that Γ ⋆,h > Γ ⋆,h−1."

    The general hierarchical decomposition is the paper's first-principles claim, yet its two load-bearing components are not derived here: Lemma 4 (strict inequality of successive minimal depths) and Theorem 3 (convergence to the trace chain at each level). Lemma 4 is exactly the condition that keeps the time scales e^{βΓ*_h} separated; without it the hierarchy is degenerate and the level-by-level Markov limits are not defined as distinct objects. The text supplies no proof for Lemma 4 or Theorem 3, and the only support invoked is [22], an arXiv preprint by the same author, which is not machine-checked or independently verified in this document. The universal assertion therefore reduces to an unexamined self-citation rather than a self-contained derivation.

  2. self citation load bearing [Section 3.4, before Theorem 9]
    "We decided to refer to [22, Section 7] for the necessary but tedious long descriptions of these stable plateaux, and just present some examples in Figure 5. For the record, the main results are as follows."

    The fourth application's entire classification of stable plateaux (Theorem 9) and the hierarchy with depths Γ*_1=1, Γ*_2=2, Γ*_3=4 (Theorem 10) are imported from [22], the same author's preprint, rather than derived or checked from independent sources in the paper. Since Theorem 10 is the concrete instance that allegedly realizes the general Theorem 3, this example cannot independently confirm the framework; it restates the same self-cited result.

full rationale

No fitted-input-called-prediction or equation-level tautology is present: the depths Γ*_h are genuine variational quantities, and the examples in Sections 3.1–3.3 cite external works (Beltrán–Landim [3,7], Kim–Seo [24], Nardi–Zocca [33], etc.). However, the paper's central universal theorem is not proved in the document. Lemma 4 and Theorem 3 are the exact structural hinge of the hierarchy, and both are deferred without proof to [22], an arXiv preprint by the same author. Under the stated rules, a self-citation is load-bearing when the central premise is justified only by that citation and the cited result is not independently verified; that is the situation here. The review nature of the paper makes the omission understandable, but the abstract's 'for any abstract lattice system' claim is supported only by the author's own full version. Hence a moderate-to-high circularity score of 6 is appropriate: the concrete examples have independent content, but the general result's derivation chain terminates in an unverified self-citation.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The central claim does not depend on fitted empirical constants. The burden is the unproved hierarchy statements, which are deferred to the author's own full version [22]. We therefore list those deferred statements as axioms of the review.

assumptions (6)
  • domain assumption Finite state space Ω, connected graph structure, and Metropolis-type transition rates (2) for all β>0.
    Section 2 opens with this setup. The entire framework, including the definition of the hierarchy and the convergence theorem, is restricted to finite connected graphs with the specific Metropolis rates.
  • ad hoc to paper Lemma 4: Γ*_h > Γ*_{h-1} for every h ≥ 2.
    Stated without proof in Section 2.1 after (28). Strict monotonicity is necessary for the time scales e^{Γ*_h β} to be distinct; if it failed, the levels of the hierarchy would not be separated.
  • ad hoc to paper Theorem 1: the number ν_h of recurrent components strictly decreases at each level.
    Stated without proof in Section 2.1. This guarantees the procedure terminates at a finite m, so the whole 'terminal level' construction depends on it.
  • ad hoc to paper Theorem 2: ground states always lie in the recurrent class at every level.
    Stated without proof in Section 2.1. It is used to identify the terminal level's unique recurrent component with the set of ground states.
  • ad hoc to paper Theorem 3: convergence of the accelerated processes to the trace Markov chains at each level.
    The main convergence result, stated without proof and deferred to the author's full version [22]. All of the model-specific theorems (4,6,8,11) are applications of this.
  • standard math The previously published results for the four Ising-model cases are correct (from [3], [24], [7], [22]).
    Section 3 states Theorems 4-11 as summaries of the cited papers; the review provides no independent verification or new proof.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Metastable Hierarchy in Abstract Low-Temperature Lattice Models." pith.science (2026). https://pith.science/paper/J7GSKAIN

@misc{pith2026250200073,
  author       = {Pith},
  title        = {Pith review of: Metastable Hierarchy in Abstract Low-Temperature Lattice Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/J7GSKAIN}},
  note         = {Machine review of arXiv:2502.00073}
}
abstract

In this article, we review the metastable hierarchy in low-temperature lattice models. In the first part, we state that for any abstract lattice system governed by a Hamiltonian potential and evolving according to a Metropolis-type dynamics, there exists a hierarchical decomposition of the collection of stable plateaux in the system into multiple $\mathfrak{m}$ levels, such that at each level there exist tunneling metastable transitions between the stable plateaux, which can be characterized by convergence to a simple Markov chain as the inverse temperature $\beta$ tends to infinity. In the second part, we collect several examples that realize this hierarchical structure of metastability. In order to fix the ideas, we select the Ising model as our lattice system and discuss its metastable behavior under four different types of dynamics, namely the Glauber dynamics with positive/zero external fields and the Kawasaki dynamics with few/many particles. This review article is submitted to the proceedings of the event PSPDE XII, held at the University of Trieste from September 9-13, 2024.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

39 extracted references · 37 canonical work pages

  1. [22]

    S. Kim. Metastable hierarchy in abstract low-temperatu re lattice models: an applica- tion to Kawasaki dynamics for Ising lattice gas with macrosc opic number of particles. arXiv:2405.08488, 2024. 1, 3.4, 9

    work page Pith review arXiv 2024

  2. [1]

    J. Ahn. Metastability of the three-state Potts model with asymmetrical external field. Stochas- tic Processes and their Applications , 176:104423, 2024. 3

    work page 2024

  3. [2]

    Beltr´ an and C

    J. Beltr´ an and C. Landim. Tunneling and metastability of continuous time Markov chains. Journal of Statistical Physics , 140(6):1065–1114, 2010. 1, 4

    work page 2010

  4. [3]

    Beltr´ an and C

    J. Beltr´ an and C. Landim. Metastability of reversible fin ite state Markov processes. Stochastic Processes and their Applications, 121(8):1633–1677, 2011. 1, 3.1, 3.1

    work page 2011

  5. [4]

    Beltr´ an and C

    J. Beltr´ an and C. Landim. Metastability of reversible co ndensed zero range processes on a finite set. Probability Theory and Related Fields , 152(3–4):781–807, 2012. 1

    work page 2012

  6. [5]

    Beltr´ an and C

    J. Beltr´ an and C. Landim. Tunneling and metastability of continuous time Markov chains II, the nonreversible case. Journal of Statistical Physics , 149(4):598–618, 2012. 1

    work page 2012

  7. [6]

    Beltr´ an and C

    J. Beltr´ an and C. Landim. A martingale approach to metast ability. Probability Theory and Related Fields, 161(1–2):267–307, 2015. 1

    work page 2015

  8. [7]

    Beltr´ an and C

    J. Beltr´ an and C. Landim. Tunneling of the Kawasaki dynam ics at low temperatures in two dimensions. Annales de l’Institut Henri Poincar´ e Probabilit´ es et Statistiques, 51(1):59–88,

    work page

Show all 39 references
  1. [8]

    Ben Arous and R

    G. Ben Arous and R. Cerf. Metastability of the three dimens ional Ising model on a torus at very low temperatures. Electronic Journal of Probability, 1:1–55, 1996. 1

    1996

  2. [9]

    G. Bet, A. Gallo and S. Kim. Metastability of the three-sta te Potts model with general inter- actions. Electronic Journal of Probability, 28:1-37, 2021. 3

    2021

  3. [10]

    G. Bet, A. Gallo and F. R. Nardi. Metastability for the deg enerate Potts model with negative ex- ternal magnetic field under Glauber dynamics. Journal of Mathematical Physics , 63:123303,

  4. [11]

    G. Bet, A. Gallo and F. R. Nardi. Metastability for the deg enerate Potts model with positive external magnetic field under Glauber dynamics. Stochastic Processes and their Applications, 172:104343, 2024. 3

    2024

  5. [12]

    Bianchi, S

    A. Bianchi, S. Dommers and C. Giardin` a. Metastability i n the reversible inclusion process. Electronic Journal of Probability, 22:1–34, 2017. 1

    2017

  6. [13]

    Bovier and F

    A. Bovier and F. den Hollander. Metastabillity: a potential-theoretic approach . Grundlehren der Mathematischen Wissenschaften 351, Springer, 2015. 1

    2015

  7. [14]

    Bovier, F

    A. Bovier, F. den Hollander and F. R. Nardi. Sharp asympto tics for Kawasaki dynamics on a finite box with open boundary. Probability Theory and Related Fields, 135(2):265–310, 2006. 1

    2006

  8. [15]

    Bovier, M

    A. Bovier, M. Eckhoff, V . Gayrard and M. Klein. Metastabi lity in reversible diffusion pro- cesses I: Sharp asymptotics for capacities and exit times. Journal of the European Mathemat- ical Society, 6(4):399–424, 2004. 1

    2004

  9. [16]

    Bovier, V

    A. Bovier, V . Gayrard and M. Klein. Metastability in reve rsible diffusion processes II: Precise asymptotics for small eigenvalues. Journal of the European Mathematical Society , 7(1):69– 99, 2005. 1

    2005

  10. [17]

    Cassandro, A

    M. Cassandro, A. Galves, E. Olivieri and M. E. V ares. Meta stable behavior of stochastic dy- namics: A pathwise approach. Journal of Statistical Physics , 35(5–6):603–634, 1984. 1

    1984

  11. [18]

    E. N. M. Cirillo, F. R. Nardi and J. Sohier. Metastability for general dynamics with rare transi- tions: Escape time and critical configurations. Journal of Statistical Physics, 161(2):365–403,

  12. [19]

    M. I. Freidlin and A. D. Wentzell. On small random perturb ations of dynamical systems. Us- pekhi Matematicheskikh Nauk, 25(1):3–55, 1970. [English translation, Russian Mathematical Surveys, 25(1):1–56, 1970.] 1

    1970

  13. [20]

    S. Kim. Second time scale of the metastability of reversi ble inclusion processes. Probability Theory and Related Fields, 180(3–4):1135–1187, 2021. 1

    2021

  14. [21]

    S. Kim. Hierarchical structure of metastability in the r eversible inclusion process: third time scale and complete characterization. arXiv:2308.13842, 2023. 1

    2023 arXiv

  15. [23]

    Kim and I

    S. Kim and I. Seo. Metastability of Ising and Potts models without external fields in large vol- umes at low temperatures. Communications in Mathematical Physics , 396(1):383–449, 2022. 1, 3

    2022

  16. [24]

    Kim and I

    S. Kim and I. Seo. Approximation method to metastability : an application to non-reversible, two-dimensional Ising and Potts models without external fie lds. arXiv:2212.13746, 2022. 1, 3.2, 3.2, 2

    2022 arXiv

  17. [25]

    Kim and I

    S. Kim and I. Seo. Energy landscape and metastability of s tochastic Ising and Potts models on three-dimensional lattices without external fields. Electronic Journal of Probability, 29:1–70,

  18. [26]

    C. Landim. Metastable Markov chains. Probability Surveys, 16:143–227, 2019. 1

    2019

  19. [27]

    Landim, J

    C. Landim, J. Lee and I. Seo. Metastability and time scale s for parabolic equations with drift 1: the first time scale. Archive for Rational Mechanics and Analysis , 248(32), 2024. 1

    2024

  20. [28]

    Landim, J

    C. Landim, J. Lee and I. Seo. Metastability and time scale s for parabolic equations with drift 2: the general time scale. arXiv:2402.07695, 2024. 1

    2024 arXiv

  21. [29]

    Landim, D

    C. Landim, D. Marcondes and I. Seo. Metastable behavior o f weakly mixing Markov chains: The case of reversible, critical zero-range processes. The Annals of Probability , 51(1):157– 227, 2023. 1

    2023

  22. [30]

    Landim, D

    C. Landim, D. Marcondes and I. Seo. A resolvent approach t o metastability. Journal of the European Mathematical Society (online), 2023. DOI 10.4171/JEMS/1398 1

    2023 doi

  23. [31]

    Landim, M

    C. Landim, M. Mariani and I. Seo. Dirichlet’s and Thomson ’s principles for non-selfadjoint elliptic operators with application to non-reversible met astable diffusion processes. Archive for Rational Mechanics and Analysis , 231(2):887–938, 2019. 1 22 Seonwoo Kim

    2019

  24. [32]

    Manzo, F

    F. Manzo, F. R. Nardi, E. Olivieri and E. Scoppola. On the e ssential features of metastability: Tunnelling time and critical configurations. Journal of Statistical Physics, 115(1–2):591–642,

  25. [33]

    F. R. Nardi and A. Zocca. Tunneling behavior of Ising and P otts models in the low-temperature regime. Stochastic Processes and their Applications , 129(11):4556–4575, 2019. 1, 1, 3.2

    2019

  26. [34]

    E. J. Neves and R. H. Schonmann. Critical droplets and met astability for a Glauber dynamics at very low temperatures. Communications in Mathematical Physics , 137(2):209–230, 1991. 1, 1, 3.1

    1991

  27. [35]

    E. J. Neves and R. H. Schonmann. Behavior of droplets for a class of Glauber dynamics at very low temperature. Probability Theory and Related Fields , 91(3–4):331–354, 1992. 1, 3.1

    1992

  28. [36]

    Olivieri and E

    E. Olivieri and E. Scoppola. Markov chains with exponent ially small transition probabilities: First exit problem from a general domain. I. The reversible case. Journal of Statistical Physics, 79(3–4):613–647, 1995. 1

    1995

  29. [37]

    Olivieri and M

    E. Olivieri and M. E. V ares. Large deviations and metastability. Encyclopedia of Mathematics and Its Applications 100, Cambridge University Press, Camb ridge, 2005. 1, 3, 2

    2005

  30. [38]

    I. Seo. Condensation of non-reversible zero-range proc esses. Communications in Mathemati- cal Physics, 366(2):781–839, 2019. 1

    2019

  31. [2022]

    3 Metastable Hierarchy in Abstract Low-Temperature Lattice Models 21

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.