Pith. sign in

REVIEW 4 minor 1 cited by

Sandpile groups of random bipartite graphs and families of distributions with the same moments

T0 review · 0 major / 4 minor · reviewed 2026-07-10 · grok-4.5

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

Pith's one-line read Large families of partition measures share the same moments, including two distinct sandpile-group laws for p=2.

desk verdict Solid construction of multi-parameter moment-sharing families that cleanly unifies the Mészáros and bipartite p=2 sandpile cases; the bipartite conjecture is open but cleanly separated. read the letter →

arxiv 2607.08607 v1 pith:7MUQG5HV submitted 2026-07-09 math.CO math.NTmath.PR

Jason Fulman , Nathan Kaplan , Deepesh Singhal , S. Ole Warnaar This is my paper · ORCID
classification math.COmath.NTmath.PR MSC 05C8005E0515B5220K01
keywords sandpilegroupsrandombipartitegraphsmethodofmomentsHall–LittlewoodfunctionsCohen–Lenstraheuristicsfiniteabelianp-groupspartitionmeasures
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

The paper builds explicit families of probability measures on partitions (equivalently, on finite abelian p-groups) that all possess identical moments, even though the measures themselves are different. These families arise naturally when one studies the Sylow p-subgroups of sandpile groups of random graphs. In particular, the special distribution that appears for even-regular graphs at p=2 and the conjectured distribution for random bipartite graphs at p=2 are distinct, yet they belong to the same one-parameter family and therefore share every moment. The construction is obtained by weighting the classical symmetric-matrix cokernel measure by monomials in the conjugate partition and then taking affine combinations of sign-twisted versions; the resulting moments are evaluated with Hall–Littlewood polynomials and basic hypergeometric identities. The work therefore shows that the method of moments, while powerful when moments grow slowly, can leave large continuous moduli of distributions once the growth rate exceeds the Wood uniqueness threshold.

What carries the argument

The measures ˜P^{Sym,κ}_{p,α} obtained by multiplying the infinite symmetric cokernel measure by p^{∑κ_i λ′_i} and then taking affine combinations of the odd-sign twists indexed by subsets of [r]; their moments are computed by expressing the surjection counts via Hall–Littlewood skew polynomials and summing the resulting products of transition kernels K(a,b) with q-hypergeometric identities.

What would settle it

Compute the empirical distribution of the 2-Sylow of the sandpile group for random bipartite graphs with part sizes n and ⌈αn⌉, α>1/2, for n several thousand, and check whether the observed frequencies for partitions of length 1–4 converge to 2^{ℓ(λ)−1}P^{Sym}_{∞,2}(λ).

Watch

Extended reading notes

Core claim

For every strict partition κ of length r there exists a 2^{r−1}-parameter family of measures on partitions, all sharing the same collection of μ-moments as the weighted measure ˜P^{Sym,κ}_p. The two special p=2 sandpile distributions—one for even-regular graphs and one conjectured for bipartite graphs—both sit inside the r=1 subfamily and therefore have identical moments p^{n(μ)+ℓ(μ)}.

Load-bearing premise

The numerical evidence offered for the bipartite-sandpile conjecture consists of only 500 samples of graphs on 100 vertices and supplies no concentration bounds that would justify the infinite-n limit.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

0 major / 4 minor

Summary. The paper constructs large families of measures on partitions that share identical μ-moments, using Hall–Littlewood specializations and transition kernels K(a,b). For a strict partition κ of length r it produces a 2^{r-1}-parameter family ˜P^{Sym,κ}_{p,α} (α∈A_r) whose moments equal those of ˜P^{Sym,κ}_p (Theorems 1.8–1.9); the r=1 subfamily recovers both Mészáros’ even-d, p=2 distribution and the conjectural p=2 bipartite distribution, both having moments p^{n(μ)+l(μ)} (Theorem 1.10). The same machinery yields closed-form moments for Cohen–Lenstra, Malle-type and Garton-type measures (Theorems 1.11–1.13). Conjecture 1.3 on Sylow p-subgroups of random bipartite graphs is stated and supported by n=100 numerics.

Significance. The work supplies the first systematic source of non-uniqueness for the method of moments when moments exceed the Wood bound, and cleanly embeds two previously isolated special cases (Mészáros even-d p=2 and the bipartite p=2 conjecture) into a single geometric family. The proofs are self-contained, relying only on the skew Cauchy identity and classical q-hypergeometric summations; the polytope A_r is shown to be a cube, guaranteeing non-negativity of the affine combinations. The moment formulae for the Cohen–Lenstra, Malle and Garton families are new for general real parameters and finite d. The numerical experiments and accompanying code repository give concrete, falsifiable predictions for Conjecture 1.3.

minor comments (4)
  1. [Section 2] Section 2: the n=100, 500-sample experiments for Conjecture 1.3 lack error bars or concentration bounds; a short remark that the numerics are only heuristic would clarify the separation between proved theorems and open conjecture.
  2. [§1.5] After (1.7): the definition of the polytope A_r is clear, but a one-sentence reminder that the inequalities are exactly those needed for non-negativity of every coordinate would help the reader.
  3. [§4.1] Lemma 4.1 and Corollary 4.2: the appeal to dominated convergence for the a o∞ limit is correct but could be flagged more explicitly for readers less familiar with q-series.
  4. [References] References: the forthcoming preprint [20] of the third author is cited for the odd-p case of Conjecture 1.3; if a public arXiv link becomes available before publication it should be added.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: moments are derived from Hall–Littlewood identities and q-series, not inserted by definition or self-citation.

full rationale

The paper’s central claims (Theorems 1.8–1.10) construct families of measures that share identical μ-moments and prove those moments equal p^{n(μ)+|κ|l(μ)} (or the normalized form p^{n(μ)+kl(μ)}). The derivation proceeds by rewriting P^Sym_∞,p via the transition kernels K(a,b) (Eqs. 4.1–4.3), expressing |Sur_p(λ,μ)| via Hall–Littlewood specializations (Prop. 1.5, Lemmas 4.3), evaluating the resulting multi-sums by induction with the q-hypergeometric identities of Lemmas 4.1, 4.4, 4.8 (proved from the 2φ0 evaluation in Prop. 6.1), and verifying that odd-cardinality sign flips leave the moments unchanged (Thm. 4.10). Normalization constants are fixed by the empty-partition moment so that the measures become probability distributions; the polytope A_r is defined by the linear inequalities that keep all probabilities non-negative (Lemma 6.5). None of these steps inserts the target moment formula by hand, fits a free parameter to data that is later “predicted,” or relies on a uniqueness theorem whose only support is a self-citation. Background formulas from Fulman–Kaplan 2019 are re-derived in Section 5 via the same Hall–Littlewood machinery. The open Conjecture 1.3 and its n=100 numerics are cleanly separated from the proved statements and do not underwrite them. Consequently the derivation chain is self-contained and free of circular reduction.

Assumptions & free parameters 0 free parameters · 4 assumptions · 1 invented entities

The paper rests on standard combinatorial and q-series machinery plus previously published universality theorems for sandpile groups; no free parameters are fitted and no new physical entities are postulated. The only non-standard inputs are the definitions of the new measures themselves, which are pure constructions.

assumptions (4)
  • domain assumption Wood’s universality theorem for cokernels of random symmetric p-adic matrices (Theorem 1.1 / [23])
    Used to identify the limiting sandpile distribution of Erdős–Rényi graphs with P^Sym_∞,p.
  • domain assumption Mészáros’s theorem for d-regular graphs (Theorem 1.2 / [15])
    Supplies the even-d, p=2 special case that is later shown to share moments with the bipartite law.
  • standard math Principal specializations and the skew Cauchy identity for Hall–Littlewood polynomials (Macdonald, Chapter III)
    Core computational engine for all moment evaluations (Sections 3–5).
  • domain assumption Koplewitz’s rank asymptotics for bipartite sandpile groups (Theorem 1.4 / [11])
    Justifies the necessity of the threshold α>1/p in Conjecture 1.3.
invented entities (1)
  • The signed measures ˜P^{Sym,κ}_{p,S} and the affine combinations ˜P^{Sym,κ}_{p,α} for α∈A_r
    purpose: To produce large families of distinct probability measures that share identical moments
    Purely combinatorial constructions; non-negativity is proved by showing A_r is a cube, but they have no independent existence outside the paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Sandpile groups of random bipartite graphs and families of distributions with the same moments." pith.science (2026). https://pith.science/paper/7MUQG5HV

@misc{pith2026260708607,
  author       = {Pith},
  title        = {Pith review of: Sandpile groups of random bipartite graphs and families of distributions with the same moments},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7MUQG5HV}},
  note         = {Machine review of arXiv:2607.08607}
}
abstract

Recently, there has been significant interest in applying the method of moments developed by Wood and others to study distributions of finite abelian groups that arise in number theory and combinatorics. When the moments do not grow too fast, they determine a unique distribution. We construct large families of distributions that have the same moments. These families include several distributions that arise naturally in the study of sandpile groups of families of random graphs. Wood determined the distribution of Sylow $p$-subgroups of sandpile groups of Erd\H{o}s--R\'enyi random graphs. This was extended by M\'esz\'aros to sandpile groups of random $d$-regular graphs, who observed an interesting special case when $d$ is even and $p = 2$. We study Sylow $p$-subgroups of sandpile groups of random bipartite graphs and similarly find a special case for $p =2$. Although this distribution differs from that of M\'esz\'aros, we show that they have the same moments and fit into our broader construction. To compute the moments of the distributions we study, we apply combinatorial tools from the theory of Hall--Littlewood functions.

Figures

Figures reproduced from arXiv: 2607.08607 by the authors.

Figure 1
Figure 1. shows the results for seven partitions λ and compares the observed frequencies with the values predicted by the conjectural distribution P Sym ∞,3 (λ). λ 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0 (1) (2) (3) (1, 1) (2, 1) (3, 1) conjecture (α, u) = (0.3, 0.5) (α, u) = (0.4, 0.5) (α, u) = (0.7, 0.2) (α, u) = (0.7, 0.5) (α, u) = (0.7, 0.8) (α, u) = (1, 0.25) (α, u) = (1, 0.5) (α, u) = (1, 0.75) [PITH_FULL_IMAGE:figures/full_fig_… view at source ↗
Figure 2
Figure 2. Comparison of the conjectural distribution P Sym,1 2 and the empirical distribution of (SGα(100,u))2 for various choices of (α, u) and λ with sample size 500. The data corresponding to α = 0.45 is far from the conjectural distribution P Sym,1 2 , which is consistent with the fact that α < 1/2 = 1/p. In contrast, for the data with α > 1/2 the empirical frequencies for these seven partitions are all close to the conje… view at source ↗
Figure 3
Figure 3. Distribution of the 2-rank of the sandpile group of a random bipartite graph. Next, we compute the expected value of p l(λ) , for a random partition λ coming from P Sym,(k) p . This is a natural quantity to consider as p rank((SGα(n,u))p) = |SGα(n,u)/pSGα(n,u) |. Lemma 2.4. Let λ be a random partition chosen from the distribution P Sym,(k) p . The expected value of p l(λ) is 1 + p k . Proof of Lemma 2.4 assuming The… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Distribution of the 3-rank of the sandpile group of random bipartite graphs. However, based on our experiments this might not always be correct. Perhaps these limits hold for β < α ⩽ 1 for some β > 1/p [PITH_FULL_IMAGE:figures/full_fig_p013_4.png]

Review history (2 revisions) →

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

  1. Distribution of Sandpile groups of random bipartite graphs

    math.CO 2026-07 accept novelty 7.0 of 10

    For odd primes p and α>1/p the p-Sylow of the sandpile group of G_α(n,u) converges in distribution to P^Sym_∞,p after discarding a rare set of graphs with too many p-divisible degrees.

Reference graph

Works this paper leans on

24 extracted references · 24 canonical work pages · cited by 1 Pith paper

  1. [1]

    Positivity preserving transformations for q-binomial coefficients

    Alexander Berkovich and S Ole Warnaar. Positivity preserving transformations for q-binomial coefficients. Transactions of the American Mathematical Society, 357(6):2291–2351, 2005

    work page 2005

  2. [2]

    The rank of the sandpile group of random directed bipartite graphs.Annals of Combinatorics, 27(4):979–992, 2023

    Atal Bhargava, Jack DePascale, and Jake Koenig. The rank of the sandpile group of random directed bipartite graphs.Annals of Combinatorics, 27(4):979–992, 2023

    work page 2023

  3. [3]

    q-Derivatives of multivariable q-hypergeometric function with respect to their parameters.Physics of Particles and Nuclei Letters, 18(3):284–289, 2021

    Vladimir V Bytev and Pengming Zhang. q-Derivatives of multivariable q-hypergeometric function with respect to their parameters.Physics of Particles and Nuclei Letters, 18(3):284–289, 2021

    work page 2021

  4. [4]

    On a Cohen–Lenstra heuristic for Jacobians of random graphs.Journal of Algebraic Combinatorics, 42(3):701–723, 2015

    Julien Clancy, Nathan Kaplan, Timothy Leake, Sam Payne, and Melanie Matchett Wood. On a Cohen–Lenstra heuristic for Jacobians of random graphs.Journal of Algebraic Combinatorics, 42(3):701–723, 2015

    work page 2015

  5. [5]

    On the distribution of divisor class groups of curves over a finite field

    Eduardo Friedman and Lawrence C Washington. On the distribution of divisor class groups of curves over a finite field. InTh´ eorie des nombres, pages 227–239. de Gruyter Berlin, 1989

    work page 1989

  6. [6]

    Random partitions and Cohen–Lenstra heuristics.Annals of Combina- torics, 23(2):295–315, 2019

    Jason Fulman and Nathan Kaplan. Random partitions and Cohen–Lenstra heuristics.Annals of Combina- torics, 23(2):295–315, 2019

    work page 2019

  7. [7]

    Sandpile groups of random bipartite graphs – numerical experiments

    Jason Fulman, Nathan Kaplan, Deepesh Singhal, and S Ole Warnaar. Sandpile groups of random bipartite graphs – numerical experiments. https://github.com/DDeepuS/Sandpile_group_Bipartite_graph, 2026. GitHub repository

    work page 2026

  8. [8]

    Random matrices, the Cohen–Lenstra heuristics, and roots of unity.Algebra & Number Theory, 9(1):149–171, 2015

    Derek Garton. Random matrices, the Cohen–Lenstra heuristics, and roots of unity.Algebra & Number Theory, 9(1):149–171, 2015

    work page 2015

Show all 24 references
  1. [9]

    Cambridge University Press, Cambridge, second edition, 2004

    George Gasper and Mizan Rahman.Basic Hypergeometric Series, volume 96 ofEncyclopedia of Mathematics and its Applications. Cambridge University Press, Cambridge, second edition, 2004

    2004

  2. [10]

    The distribution of sandpile groups of random graphs with their pairings.Trans

    Eliot Hodges. The distribution of sandpile groups of random graphs with their pairings.Trans. Amer. Math. Soc., 377(12):8769–8815, 2024

    2024

  3. [11]

    Sandpile groups of random bipartite graphs.Annals of Combinatorics, 27(1):1–18, 2023

    Shaked Koplewitz. Sandpile groups of random bipartite graphs.Annals of Combinatorics, 27(1):1–18, 2023

    2023

  4. [12]

    Cohen–Lenstra heuristics and bilinear pairings in the presence of roots of unity.arXiv:2007.12533, 2020

    Michael Lipnowski, Will Sawin, and Jacob Tsimerman. Cohen–Lenstra heuristics and bilinear pairings in the presence of roots of unity.arXiv:2007.12533, 2020

    2007 arXiv

  5. [13]

    Oxford University Press, 1998

    Ian G Macdonald.Symmetric Functions and Hall Polynomials. Oxford University Press, 1998

    1998

  6. [14]

    On the distribution of class groups of number fields.Experimental Mathematics, 19(4):465–474, 2010

    Gunter Malle. On the distribution of class groups of number fields.Experimental Mathematics, 19(4):465–474, 2010

    2010

  7. [15]

    The distribution of sandpile groups of random regular graphs.Transactions of the American Mathematical Society, 373(9):6529–6594, 2020

    Andr´ as M´ esz´ aros. The distribution of sandpile groups of random regular graphs.Transactions of the American Mathematical Society, 373(9):6529–6594, 2020

    2020

  8. [16]

    Universality for cokernels of random matrix products.Advances in Mathematics, 438:Paper No

    Hoi H Nguyen and Roger Van Peski. Universality for cokernels of random matrix products.Advances in Mathematics, 438:Paper No. 109451, 70, 2024

    2024

  9. [17]

    Local and global universality of random matrix cokernels.Math

    Hoi H Nguyen and Melanie Matchett Wood. Local and global universality of random matrix cokernels.Math. Ann., 391(4):5117–5210, 2025

    2025

  10. [18]

    Random maximal isotropic subspaces and Selmer groups.Journal of the American Mathematical Society, 25(1):245–269, 2012

    Bjorn Poonen and Eric Rains. Random maximal isotropic subspaces and Selmer groups.Journal of the American Mathematical Society, 25(1):245–269, 2012

    2012

  11. [19]

    The moment problem for random objects in a category

    Will Sawin and Melanie Matchett Wood. The moment problem for random objects in a category. arXiv:2210.06279, 2024

    2024 arXiv

  12. [20]

    Sandpile groups of random bipartite graphs.Preprint, 2026

    Deepesh Singhal. Sandpile groups of random bipartite graphs.Preprint, 2026

    2026

  13. [21]

    Symmetric functions and the explicit moment problem for abelian groups.Proc

    Roger Van Peski. Symmetric functions and the explicit moment problem for abelian groups.Proc. Amer. Math. Soc., 153(7):2799–2812, 2025

    2025

  14. [22]

    Dedekind’s η-function and Rogers–Ramanujan identities.Bulletin of the London Mathematical Society, 44(1):1–11, 2012

    S Ole Warnaar and Wadim Zudilin. Dedekind’s η-function and Rogers–Ramanujan identities.Bulletin of the London Mathematical Society, 44(1):1–11, 2012

    2012

  15. [23]

    The distribution of sandpile groups of random graphs.Journal of the American Mathematical Society, 30(4):915–958, 2017

    Melanie Matchett Wood. The distribution of sandpile groups of random graphs.Journal of the American Mathematical Society, 30(4):915–958, 2017

    2017

  16. [24]

    Probability theory for random groups arising in number theory

    Melanie Matchett Wood. Probability theory for random groups arising in number theory. InProceedings of the International Congress of Mathematicians, volume 6, pages 4476–4508, 2022. Department of Mathematics, University of Southern California Email address:fulman@usc.edu Depar...

    2022

Pith tools

Reviewed July 10, 2026 · model on record in the stance chip above.