Solomon-Terao polynomials and Castelnouvo-Mumford regularity of hyperplane arrangements

Takuro Abe

REVIEW 3 major objections 4 minor 18 references

Solomon-Terao polynomials and Castelnouvo-Mumford regularity of hyperplane arrangements

T0 review · 3 major / 4 minor · reviewed 2026-08-04 · 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 Solomon-Terao polynomial of any tame hyperplane arrangement is monic of degree equal to the number of hyperplanes.

desk verdict Settles the top-degree conjecture for Solomon-Terao polynomials; proof is mostly sound, but the acyclicity claim for multiarrangements and some coefficient formulas need fixing. read the letter →

arxiv 2509.10047 v1 pith:O7QYPRDD submitted 2025-09-12 math.AG

Takuro Abe This is my paper

classification math.AG MSC 14N2013D0252C35

keywords Solomon-TeraopolynomialhyperplanearrangementsmultiarrangementsCastelnuovo-MumfordregularitylogarithmicderivationmodulestamecharacteristicHilbertseries

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 Solomon-Terao polynomial, a specialization of the Solomon-Terao bi-polynomial, has resisted even a determination of its top degree for general arrangements. This paper settles a standing conjecture by showing that for tame arrangements the polynomial is monic of degree equal to the number of hyperplanes. The proof goes through Castelnuovo-Mumford regularity bounds for the modules of logarithmic derivations and differential forms, which control the Hilbert-series expansion that produces the polynomial. The result also holds for multiarrangements, where the degree is |m|+ℓ(d−1). These are the first general statements about the shape of Solomon-Terao polynomials outside the free case.

What carries the argument

The Castelnuovo-Mumford regularity of logarithmic derivation modules D^p(A,m) and differential forms Ω^p(A,m). The key bound is reg(D^p(A,m)) ≤ |m|−ℓ+p, proved by induction using the Derksen–Sidman approximation theorem. Combined with the Solomon-Terao complex and its acyclicity for tame arrangements, this forces the Hilbert series of the Solomon-Terao algebra to equal the bi-polynomial at t=(1−x^d)/(x−1), from which the monic top term follows.

What would settle it

Compute the Solomon-Terao polynomial of a tame multiarrangement and compare it with the Hilbert series of its Solomon-Terao algebra for a generic η: any equality failure would disprove the key acyclicity. Alternatively, find a tame arrangement where reg(D^p(A,m)) > |m|−ℓ+p for some p.

Watch

Extended reading notes

Core claim

For any essential tame multiarrangement (A,m), the order-(d+1) Solomon-Terao polynomial ST^{d+1}(A,m;x) is monic of degree |m|+ℓ(d−1). In particular, for an ordinary tame arrangement A (multiplicity 1, d=1), the Solomon-Terao polynomial ST(A;x) is monic of degree |A|, confirming Conjecture 1.6. Moreover, the coefficient of the second-highest term is ℓ+a, where a≥0 counts the degree-|m| relations in a minimal generating set of the logarithmic derivation module D^{ℓ−1}(A,m). The proof derives these facts by bounding the Castelnuovo-Mumford regularity of the logarithmic modules and then reading off the top coefficients from the bi-polynomial's Hilbert-series expression.

Load-bearing premise

The proof relies on the unstated claim, carried over from the m≡1 case, that the Solomon-Terao complex of a tame multiarrangement is acyclic outside degree zero; if that fails, the Hilbert-series identity and the monic top-degree conclusion may collapse.

Editorial extensions

If this is right

Conjecture 1.6 is true: ST(A;x) has top term x^{|A|} for every tame arrangement.
The multiarrangement version gives the same control for weighted arrangements, with degree |m|+ℓ(d−1).
For three-dimensional arrangements (which are always tame), the top-degree term of ST(A;x) is x^{|A|}.
The coefficient of x^{|A|−1} in ST(A;x) is ℓ + a with a≥0, where a is the number of degree-|A| relations in D^{ℓ−1}(A); generic arrangements achieve a = n−ℓ.
The regularity bound reg(Ω^p(A,m)) ≤ −p is a multiarrangement generalization of Saito's and Bath's results.

Reading between the lines

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

The regularity bound may apply beyond tame arrangements; the proof only needs the Solomon-Terao complex to be acyclic in positive degrees, which could hold for larger classes of (multi)arrangements.
The monic top-degree result strengthens the connection between Solomon-Terao polynomials and Poincaré polynomials of regular nilpotent Hessenberg varieties: if the degree and leading coefficient match, the polynomial is a genuine candidate for a geometric Poincaré polynomial.
A direct computational test: for a tame arrangement, the Hilbert series of the Solomon-Terao algebra S/a(A,m,η) should equal ST(A,m;x); any discrepancy would pinpoint a failure of the acyclicity assumption.
The second-highest coefficient formula gives a numerical invariant a that records syzygies of D^{ℓ−1}(A,m); it would be interesting to relate it to the intersection lattice or the matroid of A.

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

The paper proves Conjecture 1.6: for tame arrangements, the Solomon-Terao polynomial ST(A;x) is monic of degree |A|. The proof via Castelnuovo-Mumford regularity is sound; the key step in Theorem 3.7, showing the numerator of the Hilbert series has degree at most n and then extracting the leading term, is correct. The generalization to tame multiarrangements (Theorem 3.5) is genuinely new, and the second-coefficient result (Theorem 3.8) is a nice bonus.

I checked the leading term computation in Theorem 3.7. The specialization t=(1-x^d)/(x-1) turns the bi-polynomial into a sum over Hilb(D^p)(-x^d)^p. With f_ℓ=x^n and deg f_p≤n, the numerator has form (-1)^ℓ x^{n+dℓ} + ... , divided by (1-x)^ℓ. The quotient is monic of degree n+ℓ(d-1). That works.

Soft spots, in order of impact. First, Theorem 3.2 is the weakest link: acyclicity of the Solomon-Terao complex for tame multiarrangements is asserted by "the same argument as in [5]", with no proof or precise statement. If that acyclicity fails, the Hilbert-series interpretation of ST^{d+1} collapses. However, the monic-degree theorem does not actually use that interpretation — it uses the defining specialization of Ψ. So the main result is not endangered, but the paper should state a standalone acyclicity theorem for multiarrangements or cite a published one.

Second, Propositions 3.6 and 3.12 have coefficient formulas that are dimensionally impossible: dim S_i = (ℓ+i-1)!/(ℓ-1)! i!, not (ℓ+i-1)!/(ℓ-1)!. The statements as printed are wrong, though they are not used in the proof of the main theorem. That needs a correction pass.

Third, a minor typo in the proof of Theorem 3.7: the line "pd D^p ≤ ℓ-2" should be "pd Ω^{ℓ-p} ≤ ℓ-p" to yield deg f_p ≤ n. The intended argument works.

The self-citation pattern is heavy but the results cited are published; I don't see circularity. Overall, the paper deserves a serious referee. The flaws are real but localized. An editor should send it out, and the referee should ask for the corrections above.

Referee Report

3 major / 4 minor

Summary. The paper studies the Solomon–Terao bi-polynomial Ψ(A,m;x,t) of (multi)arrangements and its specialization ST(A,m;x)=Ψ(A,m;x,-1). The main claim is Theorem 3.7 (and Theorem 1.7): for an essential tame multiarrangement (A,m), the polynomial ST^{d+1}(A,m;x):=Ψ(A,m;x,(1-x^d)/(x-1)) is monic of degree |m|+ℓ(d-1), resolving Conjecture 1.6/3.4 for tame arrangements. The proof uses an inductive Castelnuovo–Mumford regularity bound (Theorem 1.10) for D^p(A,m) and Ω^p(A,m), together with tameness to control the degrees of Hilbert-series numerators. The paper also states formulas for low/high coefficients (Propositions 3.6, 3.12, Theorem 3.8) and gives examples.

Significance. If correct, the paper settles an open conjecture in a growing area connecting Solomon–Terao polynomials to Hessenberg varieties and superspace coinvariants. The main monic-degree statement is a clean, testable result, and the proof idea—using Castelnuovo–Mumford regularity of logarithmic modules—is natural and informative. The examples provide useful checks. However, the manuscript currently contains a substantial unproven acyclicity assertion (Theorem 3.2) on which the Solomon–Terao algebra interpretation rests, and two coefficient formulas are false as printed. These issues do not invalidate the central monic-degree argument for Ψ, but they require repair before the paper can be accepted.

major comments (3)

[§3, Theorem 3.2] The proof of Theorem 3.2 asserts without proof that for tame (A,m) the Solomon–Terao complex D^*(A,m) is acyclic except in degree 0, by "the same argument as in [5]". Since [5] is stated for m≡1 and no precise theorem for multiarrangements is cited, and since Definition 3.3 defines ST^{d+1} via the equality with Hilb(ST(A,m,η)), this is a load-bearing gap. If acyclicity fails, the identification with the Solomon–Terao algebra Hilbert series and the independence from η are unsupported. The proof of Theorem 3.7 uses only Ψ, so the monic-degree claim for the specialization survives; please either prove the acyclicity, cite a theorem covering the multiarrangement case, or restrict the ST^{d+1} claims accordingly.
[§3, Propositions 3.6 and 3.12] The displayed coefficient formulas are dimensionally impossible: dim_K S_i = (ℓ+i-1)!/((ℓ-1)! i!), but the formulas omit the factor 1/i!, and the x^{d+1} term omits 1/(d+1)!. For example, in Example 4.1 (ℓ=3,d=1,m≡1), Proposition 3.6 predicts a coefficient of x^2 equal to (4)!/2!=12, whereas the computed ST(A;x)=1+3x+5x^2+4x^3+x^4 has coefficient 5. The direction of the d_2 correction also appears reversed. These propositions and Corollary 3.13 are false as stated. They do not enter the proof of Theorem 3.7, but they must be corrected or withdrawn.
[§3, proof of Theorem 1.10] The induction for reg(D^p(A,m)) does not handle the subcase m(H_i)=1 and A_i:=A\{H_i} essential. Then (A,m_i) is not a multiarrangement (m_i(H_i)=0), so the previous approximation step cannot be applied, and the product decomposition using Q(A)=x_iQ(A_i) is also unavailable because A_i is essential in K^ℓ. This case can be repaired by applying the approximation theorem with M_1=D^p(A_i,m|_{A_i}) and the ideals (α_H) for H∈A, but as written the induction is incomplete. Since Theorem 1.10 supplies the degree bound deg f_p≤n used in Theorem 3.7, this gap should be filled.

minor comments (4)

[Title/Abstract] There are several typos: "Castelnouvo-Mumford" should be "Castelnuovo-Mumford", "Poinraré" should be "Poincaré", and internally "furst result" and "mulitplicity" occur. These should be corrected in revision.
[§3, Eq. (3.1)] The displayed numerator in (3.1) must be understood as (-1)^ℓ times the numerator Σ f_p(-x^d)^p; otherwise the leading sign is wrong. Please state this explicitly to avoid confusion.
[§3, proof of Theorem 3.7] The sentence "By the tameness of (A,m), it holds that pd_S D^p(A,m)≤ℓ-2" is not by itself enough to conclude deg f_p≤n. Please spell out the argument using Proposition 2.13 and tameness for Ω^{ℓ-p}: deg f_p = n + deg h_{ℓ-p}, with deg h_{ℓ-p} ≤ reg(Ω^{ℓ-p}) + pd(Ω^{ℓ-p}) ≤ -(ℓ-p)+(ℓ-p)=0.
[§3, Proposition 3.6] The proof refers to "Theorem 3.1" but the intended reference is Theorem 3.2. Also, Definition 2.16 uses η∈S^d while Theorem 3.2 uses η∈U_{d+1}; a sentence explaining that ∂ then has degree d would help readers.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the monic-degree theorem is derived directly from the definition of the Solomon-Terao bi-polynomial and independent regularity bounds, not from the target claim or from a fitted parameter.

full rationale

The central claim (Theorem 1.7 / Theorem 3.7) is that for tame (A,m), ST^{d+1}(A,m;x) = Ψ(A,m;x,(1-x^d)/(x-1)) is monic of degree |m|+ℓ(d-1). The proof in §3 computes directly from Ψ = Σ_p Hilb(D^p(A,m);x)(t(x−1)−1)^p. The only inputs are: D^ℓ(A,m) ≃ S[−n] (Prop 2.11(2)); the degree bound deg f_p ≤ n, obtained from Theorem 1.10 and tameness via Proposition 2.13; and the polynomiality of Ψ (Theorem 2.15). Substituting t=(1−x^d)/(x−1) yields a numerator whose leading term is x^{n+dℓ}, so after dividing by (1−x)^ℓ the leading term is x^{n+ℓ(d−1)}. No step fits the top degree, defines the conclusion into the input, or uses the conclusion itself. The main theorem does not rely on the Solomon-Terao algebra Hilbert series. The only appeal to prior work by the same authors is the acyclicity assertion in Theorem 3.2 ('the same argument as in [5] shows that H_p(D^*,∂^*) = 0 unless p=0'), which is used to identify Ψ with Hilb(ST(A,m,η);x) in Definition 3.3 and in the coefficient-level Propositions 3.6/3.12. That identification is not used in the proof of the monic-degree claim. Even if the multiarrangement generalization of the acyclicity were considered under-proved, that would be a correctness or completeness concern, not a circular reduction: the cited [5] result is an external published theorem for m≡1, and the paper does not define the conclusion into that hypothesis. No self-definitional, fitted-input-called-prediction, uniqueness-imported, or renaming pattern is present. Accordingly the paper receives a 0 circularity score.

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

No fitted constants or new physical entities appear. The central claim is a pure mathematical theorem resting on several prior results about tameness, logarithmic modules, and regularity, some of which are self-cited.

assumptions (6)

domain assumption The Solomon-Terao complex of a tame multiarrangement is acyclic except in degree 0, a result taken from [5].
Theorem 3.2 uses this acyclicity to rewrite Ψ as the Hilbert series of the Solomon-Terao algebra; the paper defers to 'the same argument as in [5]'.
domain assumption All 3-arrangements are tame, a fact from [3].
Used to derive Corollary 1.8 from Theorem 1.7.
domain assumption The Castelnuovo-Mumford regularity bound reg Ω^p(A) ≤ -p of Bath, Theorem 2.4.
Input for the multiarrangement regularity bounds and cited as a prior theorem.
standard math The Derksen-Sidman approximation theorem, Theorem 2.6.
Key induction step in the proof of the regularity bounds in Theorem 1.10.
standard math D^p(A,m) and Ω^p(A,m) are S-dual reflexive modules, so their projective dimensions are at most ℓ-2, from Ziegler's Proposition 2.8.
Used throughout to bound degrees of Hilbert numerators and to justify the syzygy interpretation.
domain assumption For generic η, the Solomon-Terao complex has finite-dimensional homology, Proposition 2.17 from [6].
Needed for the definition and Hilbert-series computation of the Solomon-Terao algebra.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Solomon-Terao polynomials and Castelnouvo-Mumford regularity of hyperplane arrangements." pith.science (2026). https://pith.science/paper/O7QYPRDD

@misc{pith2026250910047,
  author       = {Pith},
  title        = {Pith review of: Solomon-Terao polynomials and Castelnouvo-Mumford regularity of hyperplane arrangements},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/O7QYPRDD}},
  note         = {Machine review of arXiv:2509.10047}
}

read the original abstract

The Solomon-Terao bi-polynomial was introduced by Solomon and Terao which degenerates to the characteristic polynomial of hyperplane arrangements. Also, it was proved recently that the other specialization of the Solomon-Terao bi-polynomial, we call the Solomon-Terao polynomial, coincides with the Poinrar\'{e} polynomial of the regular nilpotent Hessenberg variety when the arrangement and the variety comes from the same lower ideal in the positive system. Moreover, there are recent developments with superspace coinvariants and Fields conjecture, thus these polynomials are becoming more and more important. However, the research of them has been very hard, and even the top degree of the Solomon-Terao polynomial has not yet been known, which we solve in this article, by using the Castelnouvo-Mumford regularity of the logarithmic derivation modules.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

18 extracted references · 4 linked inside Pith

[5]

T. Abe, T. Maeno, S. Murai and Y. Numata, Solomon-Terao alge- bra for hyperplane arrangements.J. Math. Soc. Japan71(2019), no. 4, 1027–1047

2019
[1]

Abe, Plus-one generated and next to free arrangements of hy- perplanes.Int

T. Abe, Plus-one generated and next to free arrangements of hy- perplanes.Int. Math. Res. Not2021(2021), no. 12, 9233–9261

2021
[2]

Abe, Addition-deletion theorems for the Solomon-Terao poly- nomials and B-sequences of hyperplane arrangements.Math

T. Abe, Addition-deletion theorems for the Solomon-Terao poly- nomials and B-sequences of hyperplane arrangements.Math. Z. 306(2024), no. 25

2024
[3]

Abe, Tame arrangements

T. Abe, Tame arrangements. arXiv:2504.14902

arXiv
[4]

T. Abe, T. Horiguchi, M. Masuda, S. Murai and T. Sato, Hes- senberg varieties and hyperplane arrangements.J. Reine Angew. Math.764(2020), 241–286

2020
[6]

T. Abe, H. Terao and M. Wakefield, The characteristic polynomial of a multiarrangement.Adv. in Math.,215(2007), 825–838

2007
[7]

Angarone, P

R. Angarone, P. Commins, T. Karn, S. Murai and B. Rhoades, Su- perspace coinvariants and hyperplane arrangements.Adv. Math. 467(2025), Paper No. 110185, 36 pp

2025
[8]

Bath, Hyperplane arrangements satsify (un)twisted log- arithmic comparison theorems, applications toD X-modules

D. Bath, Hyperplane arrangements satsify (un)twisted log- arithmic comparison theorems, applications toD X-modules. arXiv:2202.01462 (2022)

arXiv 2022

Show all 18 references

[9]

Derksen and J

H. Derksen and J. Sidman, Castelnuovo–Mumford regularity by approximation.Adv. Math.188(2004), 104–123

2004
[10]

Murai, B

S. Murai, B. Rhoades and A. Wilson, A proof of the Fields Con- jectures. arXiv:2505.24027 17

arXiv
[11]

Orlik and L

P. Orlik and L. Solomon, Combinatorics and topology of comple- ments of hyperplanes.Invent. Math.56(1980), no. 2, 167–189

1980
[12]

Orlik and H

P. Orlik and H. Terao,Arrangements of hyperplanes. Grundlehren der Mathematischen Wissenschaften,300. Springer-Verlag, Berlin, 1992

1992
[13]

Rose and H

L. Rose and H. Terao, A free resolution of the module of logarith- mic forms of a genericJ. Algebra136(1991), Issue 2, 376–400

1991
[14]

Saito, Degeneration of pole order spectral sequences for hyper- plane arrangements of 4-variables

M. Saito, Degeneration of pole order spectral sequences for hyper- plane arrangements of 4-variables. arxiv:1902.03838v1 (2019)

1902 arXiv
[15]

H. K. Schenck, Elementary modifications and line configurations inP 2.Comment. Math. Helv.78(2003), 447–462

2003
[16]

Solomon and H

L. Solomon and H. Terao, A formula for the characteristic poly- nomial of an arrangement.Adv. Math.64(1987), 305–325

1987
[17]

Terao, Generalized exponents of a free arrangement of hyper- planes and Shephard-Todd-Brieskorn formula.Invent

H. Terao, Generalized exponents of a free arrangement of hyper- planes and Shephard-Todd-Brieskorn formula.Invent. math.63 (1981), 159–179

1981
[18]

G. M. Ziegler, Multiarrangements of hyperplanes and their free- ness. Singularities Contemp. Math.,90, Amer. Math. Soc., Prov- idence, RI, 1989. 18

1989

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.

[1] [5]

T. Abe, T. Maeno, S. Murai and Y. Numata, Solomon-Terao alge- bra for hyperplane arrangements.J. Math. Soc. Japan71(2019), no. 4, 1027–1047

2019

[2] [1]

Abe, Plus-one generated and next to free arrangements of hy- perplanes.Int

T. Abe, Plus-one generated and next to free arrangements of hy- perplanes.Int. Math. Res. Not2021(2021), no. 12, 9233–9261

2021

[3] [2]

Abe, Addition-deletion theorems for the Solomon-Terao poly- nomials and B-sequences of hyperplane arrangements.Math

T. Abe, Addition-deletion theorems for the Solomon-Terao poly- nomials and B-sequences of hyperplane arrangements.Math. Z. 306(2024), no. 25

2024

[4] [3]

Abe, Tame arrangements

T. Abe, Tame arrangements. arXiv:2504.14902

arXiv

[5] [4]

T. Abe, T. Horiguchi, M. Masuda, S. Murai and T. Sato, Hes- senberg varieties and hyperplane arrangements.J. Reine Angew. Math.764(2020), 241–286

2020

[6] [6]

T. Abe, H. Terao and M. Wakefield, The characteristic polynomial of a multiarrangement.Adv. in Math.,215(2007), 825–838

2007

[7] [7]

Angarone, P

R. Angarone, P. Commins, T. Karn, S. Murai and B. Rhoades, Su- perspace coinvariants and hyperplane arrangements.Adv. Math. 467(2025), Paper No. 110185, 36 pp

2025

[8] [8]

Bath, Hyperplane arrangements satsify (un)twisted log- arithmic comparison theorems, applications toD X-modules

D. Bath, Hyperplane arrangements satsify (un)twisted log- arithmic comparison theorems, applications toD X-modules. arXiv:2202.01462 (2022)

arXiv 2022