A decomposition of Grassmannian associated with a hyperplane arrangement

Houshan Fu; Suijie Wang; Weikang Liang

A decomposition of Grassmannian associated with a hyperplane arrangement

T0 review · 0 major / 5 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 the A-matroid, A-adjoint, and refined A-Schubert decompositions of the Grassmannian are exactly the same for any linear hyperplane arrangement A.

desk verdict A solid extension of the essential-arrangement decomposition to all linear arrangements; the proof holds up, with only cosmetic blemishes. read the letter →

arxiv 2506.08539 v1 pith:H6EQRQTW submitted 2025-06-10 math.CO

Houshan Fu , Weikang Liang , Suijie Wang This is my paper

classification math.CO MSC 52C3505B35

keywords hyperplanearrangementadjointGrassmannianmatroiddecompositionSchubertintersectionlatticePlückercoordinatek-adjoint

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 extends a recent result from essential hyperplane arrangements to all linear hyperplane arrangements: the three ways of cutting the Grassmannian $\mathrm{Gr}(k,n)$ into pieces — by matroids, by adjoint hyperplanes, and by refined Schubert symbols — produce identical strata. This means that a single arrangement $\mathcal{A}$ induces a canonical decomposition of all $k$-dimensional subspaces of $\mathbb{R}^n$, with each piece carrying a well-defined intersection-lattice type. The main theorem, Theorem 1.3, states this equality of decompositions. As a byproduct, the paper classifies all $k$-restrictions of $\mathcal{A}$ up to intersection-lattice isomorphism, using either the matroid decomposition or the adjoint decomposition.

What carries the argument

The central mechanism is the adjoint hyperplane $H(X)$ attached to each $k$-flat $X$ of $\mathcal{A}$, defined by the signed Plücker-coordinate linear form $a_I(X)=(-1)^{\frac{1}{2}k(k+1)+\sum_{i\in I}i}\Delta_{[n]\setminus I}(X)$; these hyperplanes form the $k$-adjoint arrangement $\mathcal{A}^{(k)}$. The key identity is Lemma 3.3, which equates three conditions for a subset $I\subseteq[m]$: $I$ is a basis of $M_{\mathcal{A}}(U)$; the projected normals $\{\beta_i:i\in I\}$ are linearly independent and $|I|=k-\dim(U\cap T)$; and the flat $\bigcap_{i\in I}H_i$ is a direct complement of $U\cap(U^{\perp}+T^{\perp})$, equivalently the Plücker coordinate of the latter does not lie on the adjoint hyperplane of that flat. This dictionary converts matroid bases into adjacency in the adjoint arrangement and into dimension jumps along maximal flags, so the three decompositions refine to one.

What would settle it

Enumerate the three decompositions for a concrete non-essential central arrangement — for example, two distinct hyperplanes in $\mathbb{R}^3$ (so $T$ is a line) with $k=2$. If two $2$-planes share the same $\mathcal{A}$-matroid but fall into different $\mathcal{A}$-adjoint strata, or share an adjoint stratum but carry different Schubert symbols, the claimed equality fails. Lemma 3.3 predicts the three partitions coincide exactly; a single mismatched pair settles it.

Watch

Extended reading notes

Core claim

Let $\mathcal{A}$ be a linear hyperplane arrangement in $\mathbb{R}^n$. For each $k$-dimensional subspace $U$, three invariants are defined: the matroid $M_{\mathcal{A}}(U)$ whose rank on a subset $I$ counts the dimension of the span of the projections of the normals of the hyperplanes $H_i$ ($i\in I$) onto $U$; the stratum of the $k$-adjoint arrangement $\mathcal{A}^{(k)}$ containing the Plücker coordinate of $U\cap(U^{\perp}+T^{\perp})$; and the refined $\mathcal{A}$-Schubert symbol reading off the dimensions of $U\cap F_j$ along all maximal flags of the intersection lattice. Theorem 1.3 asserts that two subspaces coincide in one of the three corresponding decompositions if and only if they coincide in all three. The proof shows that membership in each stratum is governed by the same data: which flats $X$ of $\mathcal{A}$ complement the subspace $U\cap(U^{\perp}+T^{\perp})$, equivalently which Plücker coordinates avoid the adjoint hyperplane $H(X)$. This unifies the combinatorial, polyhedral, and Schubert perspectives for non-essential arrangements.

Load-bearing premise

All hyperplanes of $\mathcal{A}$ must pass through the origin; the proof's direct-sum and dimension arguments (Lemmas 3.1–3.3) rely on the projections $\beta_i$ and the subspace $T$ being linear, so the theorem would not hold as stated for affine arrangements.

Editorial extensions

If this is right

Every stratum of the $\mathcal{A}$-matroid decomposition has a well-defined intersection-lattice type: if $U_1$ and $U_2$ lie in the same stratum, then $L(\mathcal{A}|_{U_1})\cong L(\mathcal{A}|_{U_2})$ (Theorem 2.2).
The same classification holds through the $\mathcal{A}$-adjoint decomposition: membership in a single adjoint stratum forces isomorphic $k$-restrictions (Corollary 2.4).
When $\mathcal{A}$ is essential, the three decompositions reduce to the previously known forms, and the adjoint classification becomes: if the Plücker coordinates of two subspaces lie in the same relative interior of a flat of $\mathcal{A}^{(k)}$, their restrictions have isomorphic intersection lattices.
When $\mathcal{A}$ is the Boolean arrangement, the refined $\mathcal{A}$-Schubert decomposition is the common refinement of all $n!$ permuted Schubert decompositions, recovering the classical link between matroid strata and Schubert cells.
For a general linear arrangement with non-trivial core $T$, the strata are indexed by the dimension $i=\dim(U\cap T)$ together with flats of lower-order adjoint arrangements $\mathcal{A}^{(k-i)}$, refining the essential-case picture.

Reading between the lines

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

One could test whether the same equality survives for affine (non-linear) arrangements; the proof's dependence on orthogonal projections and direct-sum complements suggests the strata would split or merge, yielding a new invariant of affine arrangements.
The identification of matroid bases with direct complements of $U\cap(U^{\perp}+T^{\perp})$ suggests an algorithmic route to compute $M_{\mathcal{A}}(U)$ from the intersection lattice alone, independent of the chosen representative matrix.
The decomposition likely interacts with the totally nonnegative Grassmannian: when $\mathcal{A}$ is the coordinate arrangement, the strata refine the matroid strata that index positroid cells, so the general construction may yield an 'arrangement-typed' refinement of the positroid stratification.
For a family of arrangements degenerating to a non-essential one, the strata described here could track how Schubert and matroid cells merge, offering a geometric model for matroid-polytope degenerations.

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.

Desk Editor's Note

read the letter

Take it. This is a clean, honest extension of Liang–Wang–Zhao’s decomposition theorem from essential to all linear arrangements. It won’t change the way you think about Grassmannians, but it closes a gap that had to be closed, and the proof is solid.

The genuinely new piece is the index i = dim(U∩T). That forces the A-adjoint decomposition to use the (k−i)-adjoint arrangement, and the paper handles that bookkeeping correctly. Lemmas 3.1–3.3 are correct linear algebra; Lemma 3.3 is the key bridge and it checks out. The byproduct classification of k-restrictions (Theorem 2.2, Corollary 2.4) is a legitimate plus. The proof of Theorem 1.3 is self-contained modulo [6] for the essential case, which is fine—that’s a specialization, not a borrowed conclusion.

Soft spots are minor. In Lemma 3.3’s proof, they cite Theorem 3.2 where they mean Lemma 3.2; harmless but needs fixing. Definition 1.1 claims A^(n) is the empty arrangement, which is only true for non-essential arrangements; when A is essential, L_n(A) contains {0} and A^(n) has the hyperplane H({0}). The stress-test note is right that this does not change any nonempty stratum and does not affect Theorem 1.3, but the sentence itself is wrong and should be amended. The notation for the two sets L_U(A) and its complement is easy to lose in typesetting; the authors should pick more distinct symbols.

Nothing load-bearing wobbles. The proof that A-matroid and A-adjoint strata coincide is a bit dense but correct; the Schubert half is essentially immediate once Lemma 3.2 expresses dim(U∩F) in terms of the matroid rank function. No circularity, no fitted parameters, no overclaiming beyond the linear case.

This is for arrangement and matroid people who need the general statement, and for anyone who wants a clean example of extending a cellular decomposition by tracking the kernel of the arrangement. It deserves a serious referee and likely acceptance after minor revisions. I’d cite it.

Referee Report

0 major / 5 minor

Summary. The paper generalizes the decomposition theorem of Liang, Wang and Zhao from essential hyperplane arrangements to arbitrary linear hyperplane arrangements. Three decompositions of the Grassmannian are considered: the A-matroid decomposition (strata of k-subspaces having the same matroid of projected normals), the A-adjoint decomposition (strata determined by the relative interior of a flat of the k-adjoint arrangement), and the refined A-Schubert decomposition (common refinement over maximal chains of the intersection lattice). Theorem 1.3 asserts that these three decompositions are identical. The proof is direct: Lemma 3.2 provides the rank formula for the matroid M_A(U), Lemma 3.3 identifies matroid bases with intersection flats that are complementary to U∩(U⊥+T⊥), and Lemma 3.1 connects this complementarity to the adjoint hyperplanes. The paper also records a byproduct classification of k-restrictions of A (Theorem 2.2 and Corollary 2.4).

Significance. If the main theorem holds, it completes a natural extension of the GGMS-type decomposition to general linear arrangements, and it does so with a clean, essentially self-contained argument. The proof uses only linear algebra and standard lattice/matroid facts; there are no fitted parameters or assumed conclusions. The non-essential case requires the dimension parameter i=dim(U∩T), and the paper handles this correctly by passing from k-adjoints to (k-i)-adjoints. The classification byproduct is modest but useful. The main result is not surprising in view of the essential case, but the extension is nontrivial in the sense that all three definitions must be adjusted, and the proof is coherent.

minor comments (5)

[Definition 1.1] The sentence 'In particular, A^(n) is the empty arrangement' is not correct when A is essential: in that case L_n(A) contains the zero flat X={0}, and H({0}) is the hyperplane {x_∅=0} in R^{([n] n)}=R^1. The nonempty strata of the A-adjoint decomposition are unaffected because the extra flat gives an empty stratum, but the definition should be amended to avoid a false assertion.
[Lemma 3.3] In the first paragraph of the proof, the sentence 'It follows from Theorem 3.2 that dim span{β_i | i∈[m]} = k-l' should refer to Lemma 3.2; there is no Theorem 3.2 in this paper.
[Equations (3.1)-(3.2)] The two sets defined in (3.1) and (3.2) appear to differ only by an overline, which is visually easy to lose. In the proof of Theorem 1.3, the displayed identifications involving L_U(A) and ar L_U(A) should be typeset so that the overline is unmistakable; otherwise the line '∩_{X∈L_U(A)} H(X)=P' seems to contradict (3.1) unless the reader supplies the missing overline.
[Theorem 1.3] The theorem statement says 'Let A be a hyperplane arrangement A in R^n' but the whole paper assumes linear arrangements, as stated in Section 1. Adding the word 'linear' to the theorem statement would remove a potential ambiguity.
[Proof of Theorem 1.3, refined A-Schubert part] The equivalence 'X⊕(U∩(U⊥+T⊥)) = R^n if and only if there exists a flag F with σ(F)={r-k+i+1,...,r} and X=F_{r-k+i}' is correct, but it relies on the fact that complementarity forces U∩X = U∩T. Making this one-line observation explicit would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: Theorem 1.3 is proved directly from linear algebra and matroid rank identities, with prior work used only for definitions and the essential special case.

full rationale

The central claim, Theorem 1.3, is proved self-containedly in Section 3. The proof uses Lemma 3.1 (a Laplace-expansion identity relating direct-sum complements to the Plücker hyperplanes), Lemma 3.2 (a dimension formula for matroid rank in terms of subspace intersections), and Lemma 3.3 (equivalence among matroid bases, independent projections, and direct-sum complements). The A-adjoint-to-A-matroid inclusion is obtained by deriving, from the stratum condition Δ(U∩(U⊥+T⊥)) ∈ P∘ and equations (3.1)-(3.2), that LU1(A)=LU2(A) and LU1(A)=LU2(A), then applying Lemma 3.3. The converse inclusion constructs P as the common intersection of the hyperplanes H(X) for X∈LU(A) and uses the same lemmas. The A-matroid-to-refined-A-Schubert direction is proved by showing that on a refined Schubert stratum the set LU(A) is exactly the set of codimension-(k-i) flats appearing in flags with a fixed Schubert symbol, and conversely that the matroid determines all dimensions dim(U∩Fj), hence the Schubert symbol. Neither direction invokes Theorem 1.2 of [6]; the cited essential case is the specialization being generalized, not a premise. The classification byproduct in Theorem 2.2 relies on Stanley's external result [9, Proposition 3.6] about intersection lattices of arrangements, not on a conclusion of this paper. There are no fitted parameters, no data-fitting steps, and no load-bearing self-citation: [6] is used only to import definitions and to frame the extension. The minor remark that A^(n) is empty is inaccurate in the essential case because L_n(A) contains {0}, but this does not affect the nonempty strata of the A-adjoint decomposition nor the proof of Theorem 1.3, since the proof only uses L_{k-i}(A) with k-i < k when i>0 and the i=0 essential case is already covered by the definitions used in the proof. Therefore no circular step is present.

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

No free parameters and no invented entities. The paper relies only on standard matroid and arrangement theory, plus definitions inherited from [6].

assumptions (4)

standard math The intersection lattice L(A) of a linear arrangement is a geometric lattice with rank function rk(X) = n - dim(X).
Used throughout to define k-flats and maximal chains; standard theory from Orlik-Terao and Stanley.
standard math For the matroid M_A(U) represented by the projected normals beta_i, the lattice of flats is isomorphic to the intersection lattice L(A|U) of the restriction.
Invoked with Stanley [9, Proposition 3.6] to conclude in Theorem 2.2 that the same matroid stratum implies isomorphic restriction lattices.
standard math Plucker coordinates embed Gr(k,n) into projective space and satisfy Laplace expansion for stacked matrices.
Needed for Definition 1.1 of the adjoint hyperplanes and for Lemma 3.1's determinant computation.
standard math The orthogonal decomposition alpha_i = beta_i + gamma_i with beta_i in U and gamma_i in U_perp, and the rank formula dim(U intersection H_I) = k - rk_M(I).
The rank formula is proven in Lemma 3.2 from basic linear algebra and underpins Lemma 3.3 and the matroid/Schubert bridge.

how reviews work

0 comments

Cite this review

Pith. "Pith review of A decomposition of Grassmannian associated with a hyperplane arrangement." pith.science (2026). https://pith.science/paper/H6EQRQTW

@misc{pith2026250608539,
  author       = {Pith},
  title        = {Pith review of: A decomposition of Grassmannian associated with a hyperplane arrangement},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/H6EQRQTW}},
  note         = {Machine review of arXiv:2506.08539}
}

abstract

The Grassmannian, which is the manifold of all $k$-dimensional subspaces in the Euclidean space $\mathbb{R}^n$, was decomposed through three equivalent methods connecting combinatorial geometries, Schubert cells and convex polyhedra by Gelfand, Goresky, MacPherson and Serganova. Recently, Liang, Wang and Zhao discovered a novel decomposition of the Grassmannian via an essential hyperplane arrangement, which generalizes the first two methods. However, their work was confined to essential hyperplane arrangements. Motivated by their research, we extend their results to a general hyperplane arrangement $\mathcal{A}$, and demonstrate that the $\mathcal{A}$-matroid, the $\mathcal{A}$-adjoint and the refined $\mathcal{A}$-Schubert decompositions of the Grassmannian are consistent. As a byproduct, we provide a classification for $k$-restrictions of $\mathcal{A}$ related to all $k$-subspaces through two equivalent methods: the $\mathcal{A}$-matroid decomposition and the $\mathcal{A}$-adjoint decomposition.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

10 extracted references · 10 canonical work pages

[6]

$k$-Adjoint of Hyperplane Arrangements

W. Liang, S. Wang and C. Zhao.k-adjoint of hyperplane arrangements. 2025, arXiv:2412.06633v2

work page Pith review arXiv 2025
[1]

R. E. Bixby and C. R. Coullard. Adjoints of binary matroids. European J. Combin. 9 (1988), 139–147

work page 1988
[2]

B. Chen, H. Fu and S. Wang. Parallel translates of represented matroids. Adv. in Appl. Math. 127 (2021), Paper No. 102176

work page 2021
[3]

H. Fu, S. Wang. Modifications of hyperplane arrangements. J. Combin. Theory Ser. A 200 (2023), Paper No. 105797

work page 2023
[4]

W. Fulton. Young Tableaux. London Mathematical Society Student Texts, 35, Cam- bridge Univ. Press, Cambridge, 1997

work page 1997
[5]

I. M. Gel’fand, R. M. Goresky, R. D. MacPherson and V . V . Serganova. Combina- torial geometries, convex polyhedra, and Schubert cells. Adv. in Math. 63 (1987), 301–316

work page 1987
[7]

Orlik and H

P. Orlik and H. Terao. Arrangements of Hyperplanes. Grundlehren der mathematis- chen Wissenschaften, 300, Springer, Berlin, 1992

work page 1992
[8]

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

work page 2011

Show all 10 references

[9]

R. P. Stanley. An introduction to hyperplane arrangements. In: E. Miller, V . Reiner, B. Sturmfels (Eds.), Geometric Combinatorics, IAS/Park City Math. Ser. vol. 13, Amer. Math. Soc. Providence, RI, 2007, pp. 389–496

2007
[10]

H. Whitney. A logical expansion in mathematics. Bull. Amer. Math. Soc. 38 (1932), 572–579. 11

1932

Pith tools

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

[1] [6]

$k$-Adjoint of Hyperplane Arrangements

W. Liang, S. Wang and C. Zhao.k-adjoint of hyperplane arrangements. 2025, arXiv:2412.06633v2

work page Pith review arXiv 2025

[2] [1]

R. E. Bixby and C. R. Coullard. Adjoints of binary matroids. European J. Combin. 9 (1988), 139–147

work page 1988

[3] [2]

B. Chen, H. Fu and S. Wang. Parallel translates of represented matroids. Adv. in Appl. Math. 127 (2021), Paper No. 102176

work page 2021

[4] [3]

H. Fu, S. Wang. Modifications of hyperplane arrangements. J. Combin. Theory Ser. A 200 (2023), Paper No. 105797

work page 2023

[5] [4]

W. Fulton. Young Tableaux. London Mathematical Society Student Texts, 35, Cam- bridge Univ. Press, Cambridge, 1997

work page 1997

[6] [5]

I. M. Gel’fand, R. M. Goresky, R. D. MacPherson and V . V . Serganova. Combina- torial geometries, convex polyhedra, and Schubert cells. Adv. in Math. 63 (1987), 301–316

work page 1987

[7] [7]

Orlik and H

P. Orlik and H. Terao. Arrangements of Hyperplanes. Grundlehren der mathematis- chen Wissenschaften, 300, Springer, Berlin, 1992

work page 1992

[8] [8]

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

work page 2011