Pith. sign in

REVIEW 2 major objections 4 minor 31 references

Algebraic Geometry of Cactus, Pascal, and Pappus Matroids

T0 review · 2 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read This paper proves that for cactus configurations with an acyclic triple-point set, for the Pascal configuration, and for the Pappus configuration, the matroid ideal equals, up to radical, the sum of the circuit ideal, the Grassmann-Cayley…

desk verdict Real contribution, but the cactus half of the main theorem hinges on an unproved imported lemma; Pascal/Pappus also rely on black-box companion results. read the letter →

arxiv 2506.07757 v1 pith:JWSZHATZ submitted 2025-06-09 math.CO math.ACmath.AG

Emiliano Liwski , Fatemeh Mohammadi , Lisa Vandebrouck This is my paper
classification math.COmath.ACmath.AG MSC 05B3514N2014M12
keywords matroidvarietiescircuitidealscactusconfigurationsPascalconfigurationPappusGrassmann-Cayleyalgebraliftability
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

A rank-three matroid encodes which sets of points and lines in the plane are dependent. Its matroid variety is the closure of its realization space, and its defining equations - the matroid ideal - tell exactly which polynomial conditions force a configuration to be realizable. This paper proves that for cactus configurations whose triple-point set is acyclic, for the Pascal configuration, and for the Pappus configuration, the matroid ideal equals, up to radical, the sum of the circuit ideal, the Grassmann-Cayley ideal, and the lifting ideal. The result yields explicit finite generating sets for these three families, which matters because matroid ideals are notoriously hard to compute and were previously known only for much smaller examples. It also proves that every cactus configuration is realizable, that its matroid variety is irreducible, and that its circuit variety splits into at most $2^{|Q_M|}$ components obtained by turning subsets of the triple points into loops.

What carries the argument

The load-bearing objects are the matroid variety $V_M$ (the Zariski closure of the realization space), the circuit variety $V_{C(M)}$ (the locus satisfying all dependencies), and two ideals inside the matroid ideal: the Grassmann-Cayley ideal $G_M$, generated by bracket polynomials that force triples of lines to be concurrent, and the lifting ideal $I^{\mathrm{lift}}_M$, generated by minors of liftability matrices that encode when a planar configuration can be lifted from a point to a full-rank configuration in $V_{C(M)}$. The paper's key identity is $V_M = V_{C(M)} \cap V(G_M) \cap V(I^{\mathrm{lift}}_M)$; for cactus configurations the reverse inclusion is proved by the cactus perturbation lemma and a line-by-line perturbation argument, while for Pascal and Pappus it is proved by combining the irreducible decomposition of the circuit variety with the liftability criterion of Proposition 4.8.

What would settle it

Check whether Lemma 3.13, whose proof is delegated to [17, Lemma 4.23], holds for every cactus configuration with acyclic $Q_M$; a single cactus configuration with acyclic $Q_M$ and a point in $V_{C(M)} \cap V(G_M)$ that cannot be perturbed to have all $Q_M$ points nonzero would disprove Theorem 3.14. Similarly, a direct Gr\"obner-basis check that the union in Theorem 4.16 leaves out a component of the Pappus circuit variety would break Theorem 4.18.

Watch

Extended reading notes

Core claim

The central claim is that the equality $I_M = \sqrt{I_{C(M)} + G_M + I^{\mathrm{lift}}_M}$ holds for the Pascal configuration, the Pappus configuration, and every cactus configuration whose triple-point set $Q_M$ is acyclic. Here $I_{C(M)}$ is generated by brackets of the circuits, $G_M$ by Grassmann-Cayley polynomials encoding concurrence of lines, and $I^{\mathrm{lift}}_M$ by minors of liftability matrices encoding when a planar configuration can be lifted to a non-degenerate configuration in the circuit variety. The proof establishes the equivalent variety equality $V_M = V_{C(M)} \cap V(G_M) \cap V(I^{\mathrm{lift}}_M)$ and proves the reverse inclusion by perturbing arbitrary points in the intersection into the realization space. For cactus configurations the authors additionally prove realizability, irreducibility of $V_M$, and the decomposition $V_{C(M)} = \bigcup_{J \subset Q_M} V_{M(J)}$ with at most $2^{|Q_M|}$ irreducible components. This is, to the authors' knowledge, the first instance where all three constituent ideals are needed to generate a matroid ideal.

Load-bearing premise

The most fragile premise is that the cactus perturbation lemma and the Pappus circuit decomposition, imported from two companion papers, are correct; if either imported result has a gap, the corresponding main theorem inherits the gap.

Editorial extensions

If this is right

  • For the Pascal configuration, an explicit generating set up to radical consists of 7 circuit polynomials, 7 Grassmann-Cayley polynomials, and 708,588 lifting polynomials.
  • For the Pappus configuration, the analogous set has 9 circuit polynomials, 9 Grassmann-Cayley polynomials, and 2,361,960 lifting polynomials.
  • Every cactus configuration is realizable and its matroid variety is irreducible; the circuit variety has at most $2^{|Q_M|}$ irreducible components, indexed by subsets of the triple points turned into loops.
  • The Pascal circuit variety decomposes as $V_{C(N)} = V_N \cup V_{U_{2,9}} \cup \bigcup_{i=7}^9 V_{N(i)}$.
  • The equality $I_M = \sqrt{I_{C(M)} + G_M + I^{\mathrm{lift}}_M}$ provides the first instance where all three ideals are needed to describe a matroid ideal.

Reading between the lines

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

  • The enormous number of lifting generators suggests that minimal generating sets for $I^{\mathrm{lift}}_M$ are likely much smaller; the authors explicitly leave this as an open question, and testing radical membership in smaller analogues could expose redundancy.
  • The acyclicity of $Q_M$ in the cactus theorem may be a genuine boundary: Example 3.15 shows a cactus with a cyclic $Q_M$ where the simple equality fails, so an extension would require additional components or generators.
  • The same strategy - decompose the circuit variety, add concurrency and lifting conditions, then perturb - could apply to other rank-three point-line configurations whose circuit varieties have a loop-controlled irreducible decomposition, such as other grid or Steiner configurations.
  • The Pappus and Pascal generating sets depend on replacing the parameter vector $q$ by basis vectors; a different finite choice of parameters might yield drastically smaller generating sets, which would be a testable improvement.
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

2 major / 4 minor

Summary. The paper studies rank-three matroids viewed as point-line configurations and the ideals defining their matroid varieties. It introduces cactus configurations, proves that they are realizable and that their matroid varieties are irreducible, and establishes for cactus, Pascal, and Pappus configurations that the matroid ideal equals, up to radical, the sum of the circuit ideal, the Grassmann–Cayley ideal, and (for Pascal and Pappus) the lifting ideal. From these equalities it derives explicit finite generating sets for the Pascal and Pappus matroid ideals, and it gives an irreducible decomposition of the Pascal circuit variety as well as a bound on the number of irreducible components of cactus circuit varieties.

Significance. The explicit generating sets for the Pascal and Pappus matroid ideals, with the concrete generator counts in the introduction, are concrete and checkable outputs that can anchor further work on matroid ideals. The introduction of cactus configurations as a new family with realizability and irreducibility is a genuine contribution, and the paper's strategy of combining circuit, Grassmann–Cayley, and lifting ideals is clearly articulated. The internal case analyses, such as Theorem 4.13 and Proposition 4.17, are mostly explicit and give reproducible arguments.

major comments (2)
  1. [§3.2, Lemma 3.13 and Theorem 3.14] The proof of Lemma 3.13 consists solely of a reference to [17, Lemma 4.23]; the manuscript does not verify that the hypotheses of that lemma cover cactus configurations with acyclic Q_M. Since Theorem 3.14 is exactly Lemma 3.13 combined with Lemma 3.12, the cactus half of Theorem (A) has no self-contained proof. Example 3.15 shows that the acyclicity hypothesis is essential, so Lemma 3.13 is not vacuous. The authors should either prove Lemma 3.13 or include the precise statement of [17, Lemma 4.23] and a detailed verification of its hypotheses for cactus configurations.
  2. [§4.3, Theorem 4.16 and Theorem 4.18] Theorem 4.16, the irreducible decomposition of V_C(M) for the Pappus configuration, is imported from [18, §5.4] without proof and without a statement of the hypotheses. Theorem 4.18 then invokes Lemma 5.5 (ii), (iii), and (iv) of [18] in Cases 2, 3.2, 4.1, and 5, and Proposition 4.17 invokes Lemma 5.5 (iv). These lemmas are not stated in the manuscript, so the Pappus half of Theorem (A) cannot be checked from the paper alone. The authors should include the statements of these lemmas, or prove them, and verify that they apply to the Pappus configuration as defined here.
minor comments (4)
  1. [Theorem 4.18, Case 1] The sentence 'Since we must prove that γ /∈VM' should read 'γ∈V_M'; the subsequent reasoning only makes sense if the goal is to show membership in V_M.
  2. [Remark 4.19] The first sentence says 'By Theorem 4.14', but the remark is in the Pappus section and should refer to Theorem 4.18.
  3. [Lemma 3.12] In the definition of the set R, 'L_i' should be 'L_P'.
  4. [Theorem 4.14, Case 2.2] The equality 'dimγ_q(M\{7}) = dimγ_q(S_M) = 4' is confusing because S_M is defined for M, not for M\{7}; please clarify which submatroid of M\{7} is meant.

Circularity Check

3 steps flagged · score 4.0 of 10

Cactus half of Theorem (A) rests on an unproved imported lemma from a companion preprint by the same authors; the Pappus half similarly imports its circuit variety decomposition from [18].

  1. self citation load bearing [Section 3.2, Lemma 3.13 and Theorem 3.14]
    "Lemma 3.13. Let M be a cactus configuration, such that the points of Q_M do not contain a cycle. If γ∈V_C(M) ∩ V(G_M), then there exists τ∈V_C(M) such that: For every point p∈Q_M, we have τ_p ≠ 0. In particular, τ can be chosen as a perturbation of γ. Proof. The proof follows by applying the same argument as in [17, Lemma 4.23]."

    Theorem 3.14 proves I_M = sqrt(I_C(M)+G_M) by taking γ∈V_C(M)∩V(G_M), applying Lemma 3.13 to kill all zeros on Q_M, and then Lemma 3.12 to perturb to Γ_M. The only step that handles the essential acyclicity hypothesis and produces the nonzero configuration is Lemma 3.13, and that lemma is not proved in the present paper. It is outsourced to [17, Lemma 4.23], a companion preprint by the same authors. The paper does not verify that the hypotheses of [17, Lemma 4.23] cover all cactus configurations with acyclic Q_M, and Example 3.15 shows the statement is not vacuous. Thus the cactus half of the central claim reduces, in the paper's own proof chain, to an unverified self-citation.

  2. self citation load bearing [Section 3.1, Theorem 3.9]
    "Theorem 3.9. Every cactus configuration M is realizable, and its matroid variety V_M is irreducible. Proof. By Proposition 3.8, M is nilpotent, hence the result follows from [17, Theorem 4.12]."

    This is half of Theorem (B): realizability and irreducibility of V_M for every cactus configuration. The proof uses the paper's Proposition 3.8 but then immediately imports [17, Theorem 4.12], a theorem from the authors' own preprint, without proof or independent verification. Since [17, Theorem 4.12] is not machine-checked and its hypotheses are not checked against cactus configurations beyond nilpotency, the irreducibility assertion is load-bearing self-citation rather than an independent derivation within the paper.

1 more flagged steps
  1. self citation load bearing [Section 4.3, Theorem 4.16 and Theorem 4.18]
    "We first state the following result from [18,§5.4]. Theorem 4.16. The circuit variety of M admits the following irreducible decomposition V_C(M) = V_M ∪ V_U2,9 ∪ ... ∪ V_πi_M ..."

    Theorem 4.18, the Pappus part of Theorem (A), begins its case analysis from Theorem 4.16. The no-loop case of Theorem 4.18 explicitly relies on Theorem 4.16 to conclude that a configuration in V_C(M) lies in V_M, V_U2,9, or V_πi_M, and the remaining cases repeatedly invoke Lemma 5.5 of [18] (e.g., 'By Lemma 5.5 (iii) of [18], we have V_M(9)⊆V_M'). Theorem 4.16 is not proved here; it is stated as a result from [18, §5.4], a preprint by the same authors. The Pappus decomposition is therefore a load-bearing import from overlapping prior work, not a derivation established in the present paper.

full rationale

The paper is largely transparent about what it imports: Lemma 3.13 is proved by reference to [17, Lemma 4.23], Theorem 3.9 by [17, Theorem 4.12], and Theorem 4.16 by [18, §5.4]. These are companion preprints by the same authors and are not machine-checked or independently verified in the manuscript. The cactus half of Theorem (A) (Theorem 3.14) is, in the paper's own proof, exactly Lemma 3.13 followed by Lemma 3.12; the only step that handles the acyclicity condition is Lemma 3.13, so the main theorem reduces to the imported lemma. The Pappus half (Theorem 4.18) likewise cannot start without Theorem 4.16 and the repeatedly invoked Lemma 5.5 of [18], and those inclusions/decompositions are central to the case analysis. This is load-bearing self-citation, not a definitional or fitting circularity: no parameter is fitted, no quantity is renamed as a prediction, and the Pascal half (Theorem 4.13 and Theorem 4.14) is largely proved inside the paper with explicit parametrizations and case work. Since the central claim has substantial independent content but several load-bearing steps are outsourced to the authors' own unverified preprints, the appropriate score is 4 rather than 6 or higher.

Assumptions & free parameters 0 free parameters · 6 assumptions · 1 invented entities

The central claims are exact algebraic statements and contain no fitted constants. The paper does lean on a chain of imported results from the authors' own earlier work ([17] and [18]), which is the main reason the ledger lists domain assumptions rather than fully contained proofs.

assumptions (6)
  • standard math Euclidean closure of Gamma_M equals its Zariski closure over C.
    Used throughout to conclude that gamma lies in V_M from the existence of arbitrarily small perturbations in Gamma_M; standard for constructible sets over the complex numbers.
  • domain assumption Imported results from [17]: nilpotent matroids with no degree greater than two satisfy V_C(M)=V_M, solvable matroids have irreducible V_M, and V(I_lift_M) controls liftability.
    Theorems 2.7, 2.12, Proposition 2.16, and [17, Theorem 4.12] are used as black boxes; they come from a preprint by the same group and are not reproved in this paper.
  • domain assumption Lemma 3.13: for a cactus configuration M with acyclic Q_M, any gamma in V_C(M) intersected with V(G_M) can be perturbed to one with all Q_M points nonzero.
    Stated in the paper but its proof is delegated to [17, Lemma 4.23]; it is necessary for Theorem 3.14, and Example 3.15 shows the acyclicity condition cannot be dropped.
  • domain assumption Circuit variety decompositions for Pascal and Pappus, including Theorem 4.13, Theorem 4.16, and Lemma 5.5 of [18].
    These decompositions are imported from [18] and drive the ideal theorems; no proof is included in the present text.
  • domain assumption Dimension formula dim(M) and existence of a point ordering with max_i w_i at most one for nilpotent matroids.
    Used in Theorem 4.4 and in Proposition 4.8 applications to get the numerical lifting dimensions in the Pappus proof; the ordering property is asserted in Notation 4.3 from nilpotence.
  • standard math Grassmann-Cayley bracket polynomials correctly encode concurrency of lines and generate G_M without extraneous components.
    Classical exterior algebra translation; used to write explicit generators in Remarks 4.15 and 4.19.
invented entities (1)
  • Cactus configuration (rank-three matroid family)
    purpose: Central objects of the paper: point-line configurations whose associated graph is a cactus; used to state Theorems 3.9, 3.14, and 3.16.
    A new mathematical definition rather than a physical entity. It has no external falsifiable handle outside the paper, but none is needed for a combinatorial definition.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Algebraic Geometry of Cactus, Pascal, and Pappus Matroids." pith.science (2026). https://pith.science/paper/JWSZHATZ

@misc{pith2026250607757,
  author       = {Pith},
  title        = {Pith review of: Algebraic Geometry of Cactus, Pascal, and Pappus Matroids},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JWSZHATZ}},
  note         = {Machine review of arXiv:2506.07757}
}
read the original abstract

We study rank-three matroids, known as point-line configurations, and their associated matroid varieties, defined as the Zariski closures of their realization spaces. Our focus is on determining finite generating sets of defining equations for these varieties, up to radical, and describing the irreducible components of the corresponding circuit varieties. We generalize the notion of cactus graphs to matroids, introducing a family of point-line configurations whose underlying graphs are cacti. Our analysis includes several classical matroids, such as the Pascal, Pappus, and cactus matroids, for which we provide explicit finite generating sets for their associated matroid ideals. The matroid ideal is the ideal of the matroid variety, whose construction involves a saturation step with respect to all the independence relations of the matroid. This step is computationally very expensive and has only been carried out for very small matroids. We provide a complete generating set of these ideals for the Pascal, Pappus, and cactus matroids. The proofs rely on classical geometric techniques, including liftability arguments and the Grassmann--Cayley algebra, which we use to construct so-called bracket polynomials in these ideals. In addition, we prove that every cactus matroid is realizable and that its matroid variety is irreducible.

Figures

Figures reproduced from arXiv: 2506.07757 by the authors.

Figure 1
Figure 1. (Left) Three concurrent lines; (Right) Quadrilateral set. [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. (Left) Free gluing of the Fano plane and the 3 [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. (Left) The 3 × 3 grid; (Center) Fano plane; (Right) Cactus configuration. • N1 is either a line or a cycle. • For each i ∈ [k − 1], Ni+1 is the free gluing of Ni with a line or a cycle. • Nk = M. More generally, we say that a simple rank-three matroid is a cactus configuration if each of its connected components is a connected cactus configuration. An example of a cactus configuration is shown in [PITH_FULL_IMAGE:f… view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: (Left) Matroid N\{9}; (Right) Matroid A. Both arise in the proof of Lemma 4.12. Lemma 4.12. Let N be the simple rank-three matroid depicted in [PITH_FULL_IMAGE:figures/full_fig_p018_4.png]
Figure 5
Figure 5. Figure 5: Form left to right: (a) π 1 M; (b) Pascal configuration ; (c) Pascal configuration without the point 1 ;(d) Pascal configuration without the point 7. It follows that the vectors {γ1, . . . , γ9} all lie in a common two-dimensional subspace. Consequently, we conclude th…
Figure 6
Figure 6. Figure 6: (Left) Pappus Configuration, (Center) I9, (Right) J1 Case 1. There exists p ∈ [8]\{1, 4} such that γl1 ∧ γl2 ̸= 0, where {l1, l2} = Lp. In this case, since N\{p} is nilpotent and all its points have degree at most two, Theorem 2.7 (i) allows us to perturb the collectio…
Figure 7
Figure 7. Figure 7: (Left) π i M; (Center) M(9); (Right) M(1, 9). is a full-rank submatroid of M, Theorem 2.12 allows us to infinitesimally lift the vectors {γ2, . . . , γ9} from γ1, to obtain vectors {γe2, . . . , γe9} ∈ VC(M\{1}) with rank{γ2, . . . , γ9} = 3. Letting γe denote the resu…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

31 extracted references · 30 canonical work pages

  1. [17]

    Liwski and F

    E. Liwski and F. Mohammadi. Paving matroids: defining equations and associated varieties. arXiv preprint arXiv:2403.13718, 2024

    work page arXiv 2024

  2. [18]

    Minimal matroids in dependency posets: algorithms and applications to computing irreducible decompositions of circuit varieties

    E. Liwski and F. Mohammadi. Minimal matroids in dependency posets: algorithms and applications to computing irreducible decompositions of circuit varieties.arXiv preprint arXiv:2502.00799, 2025

    work page Pith review arXiv 2025

  3. [1]

    Bruns and A

    W. Bruns and A. Conca. Gr¨ obner bases and determinantal ideals. InCommutative Algebra, Singularities and Computer Algebra, pages 9–66. Springer Netherlands, 2003

    work page 2003

  4. [2]

    Caines, F

    P. Caines, F. Mohammadi, E. S´ aenz-de Cabez´ on, and H. Wynn. Lattice conditional independence models and Hibi ideals.Transactions of the London Mathematical Society, 9(1):1–19, 2022

    work page 2022

  5. [3]

    Clarke, K

    O. Clarke, K. Grace, F. Mohammadi, and H. Motwani. Matroid stratifications of hypergraph varieties, their realization spaces, and discrete conditional independence models.International Mathematics Research Notices, page rnac268, 2022

    work page 2022

  6. [4]

    Clarke, G

    O. Clarke, G. Masiero, and F. Mohammadi. Liftable point-line configurations: Defining equations and irreducibility of associated matroid and circuit varieties.Mathematics, 12(19):3041, 2024

    work page 2024

  7. [5]

    Clarke, F

    O. Clarke, F. Mohammadi, and H. Motwani. Conditional probabilities via line arrangements and point configurations.Linear and Multilinear Algebra, 70(20):5268–5300, 2022

    work page 2022

  8. [6]

    Drton, B

    M. Drton, B. Sturmfels, and S. Sullivant.Lectures on Algebraic Statistics, volume 39. Birkh¨ auser, Basel, first edition, 2009

    work page 2009

Show all 31 references
  1. [7]

    V. Ene, J. Herzog, T. Hibi, and F. Mohammadi. Determinantal facet ideals.Michigan Mathematical Journal, 62(1):39–57, 2013

    2013

  2. [8]

    L. M. Feh´ er, A. N´ emethi, and R. Rim´ anyi. Equivariant classes of matrix matroid varieties. Commentarii Mathematici Helvetici, 87(4):861–889, 2012

    2012

  3. [9]

    Gelfand, M

    I. Gelfand, M. Goresky, R. MacPherson, and V. Serganova. Combinatorial geometries, convex polyhedra, and schubert cells.Advances in Mathematics, 63(3):301–316, 1987

    1987

  4. [10]

    J. E. Graver, B. Servatius, and H. Servatius.Combinatorial rigidity. Number 2 in Mathematical Sciences Series. American Mathematical Soc., 1993

    1993

  5. [11]

    Guerville-Ball´ e and J

    B. Guerville-Ball´ e and J. Viu-Sos. Connectedness and combinatorial interplay in the moduli space of line arrangements, 2023

    2023

  6. [12]

    Hartshorne.Algebraic geometry, volume 52

    R. Hartshorne.Algebraic geometry, volume 52. Springer Science & Business Media, 2013. 26

    2013

  7. [13]

    Herzog, T

    J. Herzog, T. Hibi, F. Hreinsd´ ottir, T. Kahle, and J. Rauh. Binomial edge ideals and conditional independence statements.Advances in Applied Mathematics, 3(45):317–333, 2010

    2010

  8. [14]

    Ho¸ sten and S

    S. Ho¸ sten and S. Sullivant. Ideals of adjacent minors.Journal of Algebra, 277(2):615–642, 2004

    2004

  9. [15]

    Jackson and S.-i

    B. Jackson and S.-i. Tanigawa. Maximal matroids in weak order posets.Journal of Combinatorial Theory, Series B, 165:20–46, 2024

    2024

  10. [16]

    Knutson, T

    A. Knutson, T. Lam, and D. E. Speyer. Positroid varieties: juggling and geometry.Compositio Mathematica, 149(10):1710–1752, 2013

    2013

  11. [19]

    Liwski, F

    E. Liwski, F. Mohammadi, and R. Pr´ ebet. Efficient algorithms for minimal matroid extensions and irreducible decompositions of circuit varieties.arXiv preprint arXiv:2504.16632, 2025

    2025

  12. [20]

    Oxley.Matroid Theory

    J. Oxley.Matroid Theory. Second edition, Oxford University Press, 2011

    2011

  13. [21]

    Pfister and A

    G. Pfister and A. Steenpass. On the primary decomposition of some determinantal hyperedge ideal.Journal of Symbolic Computation, 103:14–21, 2019

    2019

  14. [22]

    M. J. Piff and D. J. Welsh. On the vector representation of matroids.Journal of the London Mathematical Society, 2(2):284–288, 1970

    1970

  15. [23]

    Poljak and D

    S. Poljak and D. Turz ´ ık. Amalgamation over uniform matroids.Czechoslovak Mathematical Journal, 34(2):239–246, 1984

    1984

  16. [24]

    Sidman, W

    J. Sidman, W. Traves, and A. Wheeler. Geometric equations for matroid varieties.Journal of Combinatorial Theory, Series A, 178:105360, 2021

    2021

  17. [25]

    Sitharam and A

    M. Sitharam and A. Vince. The maximum matroid of a graph.arXiv preprint arXiv:1910.05390

    1910 arXiv

  18. [26]

    Studen´ y.Probabilistic conditional independence structures

    M. Studen´ y.Probabilistic conditional independence structures. Springer, London, 2005

    2005

  19. [27]

    Sturmfels

    B. Sturmfels. On the matroid stratification of Grassmann varieties, specialization of coordinates, and a problem of N. White.Advances in Mathematics, 75(2):202–211, 1989

    1989

  20. [28]

    Sturmfels.Algorithms in invariant theory

    B. Sturmfels.Algorithms in invariant theory. Springer-Verlag, Berlin, Heidelberg, 1993

    1993

  21. [29]

    Vakil.The Rising Sea, Foundations of Algebraic Geometry

    R. Vakil.The Rising Sea, Foundations of Algebraic Geometry. Available at http://math.stanford. edu/∼vakil/216blog/FOAGnov1817public, 2017

    2017

  22. [30]

    Whiteley

    W. Whiteley. Some matroids from discrete applied geometry.Contemporary Mathematics, 197:171–312, 1996

    1996

  23. [31]

    H. Whitney. On the abstract properties of linear dependence.Amer. J. Math., 57(3):509, jul 1935. Authors’ addresses Emiliano Liwski, KU Leuvenemiliano.liwski@kuleuven.be Fatemeh Mohammadi, KU Leuvenfatemeh.mohammadi@kuleuven.be Lisa Vandebrouck, KU Leuvenlisa.vandebrouck@stude...

    1935

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.