pith. sign in

arxiv: 2304.07405 · v2 · submitted 2023-04-14 · 🧮 math.AG · math.CO

Bounding the number of graph refinements for Brill-Noether existence

Karl Christ , Qixiao Ma This is my paper

Pith reviewed 2026-05-24 09:28 UTC · model grok-4.3

classification 🧮 math.AG math.CO
keywords Brill-Noether theorygraph divisorshomothetic refinementalgebraic curvesspecial divisorsgenus grank and degree
0
0 comments X

The pith

Algebraic geometry results yield an explicit upper bound on the refinements needed for Brill-Noether divisors to exist on graphs.

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

The paper shows that for any finite graph G of genus g, and integers d and r with non-negative Brill-Noether number, there is an explicit upper bound K(g,d,r) such that the K-th homothetic refinement of G has a divisor of degree d and rank at least r. This turns the known existence of some large enough k into a concrete, computable bound derived from algebraic geometry theorems. A reader would care because it makes the result effective rather than purely existential, bridging combinatorial models of curves with classical theory. The bound depends only on the numerical invariants g, d, and r, not on the specific structure of G.

Core claim

Using results from algebraic geometry, the authors produce an explicit upper bound for k in terms of g, d, and r, ensuring that the k-th homothetic refinement G^{(k)} admits a divisor of degree d and rank at least r whenever the Brill-Noether number is non-negative.

What carries the argument

The k-th homothetic refinement of the graph, combined with the application of algebraic geometry theorems on the existence of special divisors on curves.

If this is right

  • For every graph G of genus g and every d, r with ρ(g,d,r) ≥ 0, the refinement G^{(k)} has the desired divisor for all k at least the given bound.
  • The bound is finite and depends only on g, d, r.
  • This provides a uniform way to guarantee Brill-Noether existence across all graphs of a given genus.
  • Existence becomes effective: one can in principle construct the divisor after a known number of refinements.

Where Pith is reading between the lines

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

  • One could use the bound to run computer searches for small g to find actual minimal k for specific graphs.
  • The result suggests that Brill-Noether theory on graphs can be made quantitative by lifting to algebraic curves.
  • Similar bounding techniques might apply to other existence questions in graph divisor theory.

Load-bearing premise

That known algebraic-geometry theorems can be applied directly to the homothetic refinements of an arbitrary finite graph to produce an explicit, finite upper bound on k.

What would settle it

Finding a graph G of genus g together with d and r where the Brill-Noether number is non-negative, but the minimal k required for G^{(k)} to have a divisor of degree d and rank r exceeds the upper bound provided by the paper.

Figures

Figures reproduced from arXiv: 2304.07405 by Karl Christ, Qixiao Ma.

Figure 1
Figure 1. Figure 1: On the left, a graph G that does not admit a g 1 5 . On the right, the first homothetic refinement G(1), together with a g 1 5 . 2 2 2 2 2 Γ Σ v [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: On the left, Γ and Σ. On the right, the map Γ → Σ zoomed in at a subgraph of Γ; the map is defined analogously on the other parts of Γ. Numbers indicate the degree of the map on a vertex, if it is different from 1; half-edges indicate branch- or ramification points. The map has stretching factor 1 along all edges and edges in the same fiber have the same colour. The blue and brown edges have half the lengt… view at source ↗
read the original abstract

Let $G$ be a finite graph of genus $g$. Let $d$ and $r$ be non-negative integers such that the Brill-Noether number is non-negative. It is known that for some $k$ sufficiently large, the $k$-th homothetic refinement $G^{(k)}$ of $G$ admits a divisor of degree $d$ and rank at least $r$. We use results from algebraic geometry to give an upper bound for $k$ in terms of $g,d,$ and $r$.

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.

Referee Report

2 major / 2 minor

Summary. The paper claims that for any finite graph G of genus g, and nonnegative integers d,r with Brill-Noether number ρ(g,r,d) ≥ 0, there exists an explicit upper bound K = K(g,d,r) (independent of the particular graph G) such that the k-th homothetic refinement G^(k) admits a divisor of degree d and rank at least r for all k ≥ K. The bound is obtained by appealing to known existence theorems for linear series on algebraic curves and transferring them to the graph setting via the skeleton construction.

Significance. If the claimed bound is indeed independent of G and fully explicit, the result would make the 'sufficiently large k' in Brill-Noether theory for graphs uniform and computable from g,d,r alone, strengthening the bridge between combinatorial graph theory and algebraic geometry. This could facilitate effective versions of tropical Brill-Noether theory and related applications in moduli spaces.

major comments (2)
  1. [Abstract and main theorem statement] The central claim that the upper bound on k depends only on g,d,r (and not on the combinatorics of G) rests on the assumption that, for any G, one can always find a curve with skeleton exactly G^(k) that is sufficiently general in the moduli space to apply the algebraic-geometry existence theorem. This step is not immediate, because generality may require metric choices or positions that implicitly depend on the number of vertices/edges of G or on the refinement level in a non-uniform way. The proof must exhibit an explicit construction or citation showing independence from G.
  2. [Proof of the main result] The transfer from the algebraic curve to the graph G^(k) via the skeleton map requires that the linear series on the curve specializes to a divisor of the claimed rank on the graph. The manuscript should verify that the Brill-Noether existence result used produces a series whose specialization is guaranteed to have rank at least r on every possible G^(k), without additional constraints on k that depend on G.
minor comments (2)
  1. [Introduction] Clarify the precise statement of the Brill-Noether number ρ(g,r,d) used and confirm it matches the standard definition.
  2. [Preliminaries] The notation for the homothetic refinement G^(k) should be defined explicitly, including how the edge lengths are scaled.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and the detailed comments. We address the two major comments point by point below, providing clarifications on the uniformity of the bound K(g,d,r). We will make the requested explicit justifications in a revised version of the manuscript.

read point-by-point responses

  1. Referee: [Abstract and main theorem statement] The central claim that the upper bound on k depends only on g,d,r (and not on the combinatorics of G) rests on the assumption that, for any G, one can always find a curve with skeleton exactly G^(k) that is sufficiently general in the moduli space to apply the algebraic-geometry existence theorem. This step is not immediate, because generality may require metric choices or positions that implicitly depend on the number of vertices/edges of G or on the refinement level in a non-uniform way. The proof must exhibit an explicit construction or citation showing independence from G.

    Authors: The uniformity of K follows from the fact that M_g is irreducible of fixed dimension 3g-3 and the Brill-Noether locus has expected codimension independent of any graph. For any metric graph, including G^(k) with edge lengths scaled by 1/k, the preimage under the tropicalization map contains an open set of curves; choosing the metric generically ensures the curve lies outside the Brill-Noether special locus. The minimal k making all edge lengths sufficiently small to enter this open set is bounded by a constant depending only on g (via the degree of the relevant divisor class on M_g), independent of the initial combinatorics of G. We will add a paragraph in Section 2 with a citation to the dominance of the tropicalization map (e.g., to results in the literature on moduli of curves with prescribed dual graph) to make this explicit. revision: yes

  2. Referee: [Proof of the main result] The transfer from the algebraic curve to the graph G^(k) via the skeleton map requires that the linear series on the curve specializes to a divisor of the claimed rank on the graph. The manuscript should verify that the Brill-Noether existence result used produces a series whose specialization is guaranteed to have rank at least r on every possible G^(k), without additional constraints on k that depend on G.

    Authors: By the specialization lemma, any linear series of rank r on the curve specializes to a divisor of rank at least r on its skeleton. The algebraic existence theorem supplies such a series on a general curve; once k is large enough for G^(k) to arise as the skeleton of a general curve (as justified in the response to the first comment), the specialization automatically yields rank at least r on G^(k). No further G-dependent lower bound on k appears, because the generality condition is open in M_g and the required refinement scale is controlled uniformly by the fixed dimension of M_g. We will insert a short verification paragraph immediately after the statement of the specialization lemma, referencing Baker's lemma and confirming the absence of G-dependent constraints. revision: yes

Circularity Check

0 steps flagged

No circularity: bound imported from external AG theorems

full rationale

The paper's central claim is that known Brill-Noether existence theorems from algebraic geometry can be applied to sufficiently refined graphs G^(k) to produce an explicit upper bound on k depending only on g, d, r. The abstract explicitly frames the contribution as importing and applying external AG results rather than deriving the bound internally from graph combinatorics or fitting parameters. No self-definitional steps, fitted inputs renamed as predictions, or load-bearing self-citations appear in the derivation chain. The result is therefore self-contained against external benchmarks and receives the default non-circularity finding.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only; no free parameters, axioms, or invented entities are described.

pith-pipeline@v0.9.0 · 5605 in / 995 out tokens · 25439 ms · 2026-05-24T09:28:47.324599+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

16 extracted references · 16 canonical work pages

  1. [1]

    Lifting harmonic morphisms I : metrized complexes and B erkovich skeleta

    Omid Amini, Matthew Baker, Erwan Brugall\' e , and Joseph Rabinoff. Lifting harmonic morphisms I : metrized complexes and B erkovich skeleta. Res. Math. Sci. , 2:Art. 7, 67, 2015

    work page 2015

  2. [2]

    Arbarello, M

    E. Arbarello, M. Cornalba, P. A. Griffiths, and J. Harris. Geometry of algebraic curves. V ol. I , volume 267 of Grundlehren der mathematischen Wissenschaften [Fundamental Principles of Mathematical Sciences] . Springer-Verlag, New York, 1985

    work page 1985

  3. [3]

    A note on B rill- N oether existence for graphs of low genus

    Stanislav Atanasov and Dhruv Ranganathan. A note on B rill- N oether existence for graphs of low genus. Michigan Math. J. , 67(1):175--198, 2018

    work page 2018

  4. [4]

    Specialization of linear systems from curves to graphs

    Matthew Baker. Specialization of linear systems from curves to graphs. Algebra Number Theory , 2(6):613--653, 2008. With an appendix by Brian Conrad

    work page 2008

  5. [5]

    Degeneration of linear series from the tropical point of view and applications

    Matthew Baker and David Jensen. Degeneration of linear series from the tropical point of view and applications. In Nonarchimedean and tropical geometry , Simons Symp., pages 365--433. Springer, [Cham], 2016

    work page 2016

  6. [6]

    Baker and S

    M. Baker and S. Norine. Riemann- R och and A bel- J acobi theory on a finite graph. Adv. Math. , 215(2):766--788, 2007

    work page 2007

  7. [7]

    Barth, C

    W. Barth, C. Peters, and A. Van de Ven. Compact complex surfaces , volume 4 of Ergebnisse der Mathematik und ihrer Grenzgebiete (3) [Results in Mathematics and Related Areas (3)] . Springer-Verlag, Berlin, 1984

    work page 1984

  8. [8]

    Algebraic and combinatorial B rill- N oether theory

    Lucia Caporaso. Algebraic and combinatorial B rill- N oether theory. In Compact moduli spaces and vector bundles , volume 564 of Contemp. Math. , pages 69--85. Amer. Math. Soc., Providence, RI, 2012

    work page 2012

  9. [9]

    Tropicalizing the space of admissible covers

    Renzo Cavalieri, Hannah Markwig, and Dhruv Ranganathan. Tropicalizing the space of admissible covers. Math. Ann. , 364(3-4):1275--1313, 2016

    work page 2016

  10. [10]

    Brill- N oether conjecture on cactus graphs

    Phan Thi Ha Duong. Brill- N oether conjecture on cactus graphs. Acta Math. Vietnam. , 47(4):833--845, 2022

    work page 2022

  11. [11]

    Algebraic geometry

    Robin Hartshorne. Algebraic geometry . Graduate Texts in Mathematics, No. 52. Springer-Verlag, New York-Heidelberg, 1977

    work page 1977

  12. [12]

    Rank of divisors on tropical curves

    Jan Hladk\' y , Daniel Kr\' a l , and Serguei Norine. Rank of divisors on tropical curves. J. Combin. Theory Ser. A , 120(7):1521--1538, 2013

    work page 2013

  13. [13]

    On the K odaira dimension of the moduli space of curves

    Joe Harris and David Mumford. On the K odaira dimension of the moduli space of curves. Invent. Math. , 67(1):23--88, 1982. With an appendix by William Fulton

    work page 1982

  14. [14]

    Lectures on resolution of singularities , volume 166 of Annals of Mathematics Studies

    J\' a nos Koll\' a r. Lectures on resolution of singularities , volume 166 of Annals of Mathematics Studies . Princeton University Press, Princeton, NJ, 2007

    work page 2007

  15. [15]

    Brill- N oether existence on graphs via R -divisors, polytopes and lattices

    Madhusudan Manjunath. Brill- N oether existence on graphs via R -divisors, polytopes and lattices. Selecta Math. (N.S.) , 28(2):Paper No. 35, 26, 2022

    work page 2022

  16. [16]

    Discrete and metric divisorial gonality can be different

    Josse van Dobben de Bruyn, Harry Smit, and Marieke van der Wegen. Discrete and metric divisorial gonality can be different. J. Combin. Theory Ser. A , 189:Paper No. 105619, 19, 2022

    work page 2022