On modular inequalities for plane projective curves
Jose Ignacio Cogolludo-Agustín ∗, Anca Măcinic ∗∗ · 2026
Read On modular inequalities for plane projective curves by Jose Ignacio Cogolludo-Agustín ∗, Anca Măcinic ∗∗ (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-26237.txt.
Introduction
Cogolludo-Agustín and Măcinic introduce inequalities, called modular, for complements of plane curves, built from a combinatorial Aomoto complex tied to a curve's weak type. They use the tool to study twisted Alexander polynomials, especially the characteristic polynomial of the Milnor fiber of a projective plane curve. They give criteria for nontrivial resonance in degree one over positive-characteristic fields for quasi fiber-type curves, plus lower bounds for twisted root multiplicities. They apply the results to compute root multiplicities for fiber-type curves.
Catalog record
| CREATOR | Jose Ignacio Cogolludo-Agustín ∗, Anca Măcinic ∗∗ |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 93288 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.26237
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
On modular inequalities for plane projective curves Thanks: ∗ Partially supported by a grant of the Spanish Ministerio de Ciencia, Innovación y Universidades MICIU/AEI/10.13039/501100011033/FEDER, UE, reference PID2024-156181NB-C33. Thanks: ∗∗ Partially supported by a grant of the Ministry of Research, Innovation and Digitization, CNCS - UEFISCDI, project number PN-IV-P1-PCE-2023-2001, within PNCDI IV Jose Ignacio Cogolludo-Agustín ∗, Anca Măcinic ∗∗ Abstract . We introduce modular inequalities for complements of plane curves, based on a Combinatorial Aomoto complex construction associated with the weak combinatorial type of a curve. We use this as a tool to investigate twisted Alexander polynomials, in particular to study the characteristic polynomial of the Milnor fiber associated to a projective plane curve, i.e. the classical Alexander polynomial. We give criteria for the non-triviality of the resonance in degree 1 1 , over a positive characteristic field, for curves with quasi fiber–type structure and we compute lower bounds for the multiplicities of some roots of twisted Alexander polynomials for this type of curves. We apply these results to theoretically compute the multiplicities of certain roots of twisted Alexander polynomials for fiber–type curves. Mathematics Subject Classification (2020) 14H50, 32S55, 32S40, 14C21, 14Q05, 14B05 1. Introduction The Milnor fiber \mathcal{F}(\mathcal{C}) of a complex projective curve \mathcal{C}\subset\mathbb{P}^{2} , defined by a homogeneous polynomial f of degree d , is the smooth affine hypersurface \mathcal{F}(\mathcal{C})=\{(x,y,z)\in\mathbb{C}^{3}\setminus\{0\}\mid f(x,y,z)=1\} . The action of a d -th root of unity over \mathcal{F}(\mathcal{C}) by h(x,y,z)=(\xi x,\xi y,\xi z) endows the homology H_{1}(\mathcal{F}(\mathcal{C});\mathbb{Q}) with a \mathbb{Q}[t^{\pm 1}] –module structure. Since h_{*} has finite order d , this is a semi-simple torsion module. Moreover, since \mathbb{Q}[t^{\pm 1}] is a PID each of its torsion components is generated by a (Laurent) polynomial associated with \phi_{k} , where \phi_{k}(t) is a cyclotomic polynomial of order k|d . The product of all these polynomials (1) \Delta_{\mathcal{C}}(t)=\prod_{k|d}\phi_{k}^{b_{k}}(t) is referred to as the Alexander polynomial of \mathcal{C} . The module H_{1}(\mathcal{F}(\mathcal{C});\mathbb{Q}) controls the topology of a d -cyclic cover of the complement \mathbb{P}^{2}\setminus\mathcal{C} ramifying with maximal order at every irreducible component of \mathcal{C} , that is, associated with the epimorphism \varepsilon:H_{1}(\mathbb{P}^{2}\setminus\mathcal{C})\to\mathbb{Z}_{d} defined as \varepsilon(\mu_{i})=1 for any meridian \mu_{i} around an irreducible component \mathcal{C}_{i} of \mathcal{C} . If one is interested in cyclic d -covers of \mathbb{P}^{2}\setminus\mathcal{C} associated with other epimorphisms \varepsilon:H_{1}(\mathbb{P}^{2}\setminus\mathcal{C})\to\mathbb{Z}_{d} defined as \varepsilon(\mu_{i})=m_{i}\in\mathbb{Z}_{d} it is useful to define the twisted Milnor fiber \mathcal{F}(\mathcal{C},\bar{m})=\{(x,y,z)\in\mathbb{C}^{3}\setminus\{0\}\mid F_{1}^{m_{1}}\cdots F_{r}^{m_{r}}(x,y,z)=1\} associated with the weights \bar{m}=(m_{1},\dots,m_{r}) as long as \gcd(\bar{m})=1 , which is the condition for \varepsilon to be an epimorphism. The torsion of the \mathbb{Q}[t^{\pm 1}] –module H_{1}(\mathcal{F}(\mathcal{C},\bar{m});\mathbb{Q}) also defines the twisted Alexander polynomial (2) \Delta_{(\mathcal{C},\bar{m})}(t)=\prod_{k|d}\phi_{k}^{b_{k,\bar{m}}}(t). For complex projective line arrangements \mathcal{L} , Papadima–Suciu [ 20 ] provide combinatorial upper bounds for b_{k} in ( 1 ) for the Alexander polynomial \Delta_{\mathcal{L}}(t) and k a prime power. In the theory of arrangements of hyperplanes, the combinatorics of an arrangement reverberates into the topological and algebraic properties of the arrangement and of its complement. Let us mention, for instance, the emblematic result by Brieskorn [ 3 ] and completed by Orlik–Solomon in [ 19 ] which states that the cohomology algebra of the complement of an arrangement of hyperplanes with integer coefficients is completely determined by the intersection lattice of the arrangement. The purpose of this paper is to give estimates for the multiplicities b_{k} of the classical Alexander polynomial in ( 1 ) and b_{k,\bar{m}} for the twisted version in ( 2 ) in case k=p^{s} , for p prime. In order to do that we use two different tools that offer respectively upper and lower bounds to these multiplicities b_{k} and b_{k,\bar{m}} . Upper bounds are obtained via the weak combinatorial type of a curve \mathcal{C} , which can be seen as a generalization to curves of the classical concept of combinatorics of line arrangements, developed by the first author in [ 4 ] and jointly with Matei in [ 5 ] . The authors prove that the weak combinatorial type of a curve determines its cohomology ring. The cohomology ring over a finite field \Bbbk produces a complex given by multiplication by a 1-form \omega\in H^{1}(X;\Bbbk) denoted by (H^{*}(X;\Bbbk),\cdot\wedge\omega) and called the Aomoto Complex . Note that the cohomology of these complexes produces a filtration of H^{1}(X;\Bbbk) called the resonance varieties of X and defined as {\mathcal{R}}_{k}(X,\Bbbk):=\{\omega\in H^{1}(X;\Bbbk)\mid\dim_{\Bbbk}H^{1}(H^{*}(X;\Bbbk),\cdot\wedge\omega)\geq k\}. On the other hand, given the weak combinatorial type of a curve we define in § 3 a graded algebra A_{W,\Bbbk}^{*} over a field \Bbbk . Multiplication by an element of order one \omega\in A_{W,\Bbbk}^{1} also produces a family of complexes called Combinatorial Aomoto Complex and denoted by (A_{W,\Bbbk}^{1},\cdot\curlywedge\omega) whose cohomology produces a filtration of A_{W,\Bbbk}^{1} called the combinatorial resonance varieties of W and defined as {\mathcal{R}}_{k}(W,\Bbbk):=\{\omega\in A_{W,\Bbbk}^{1}\mid\dim_{\Bbbk}H^{1}(A_{W,\Bbbk}^{*},\cdot\curlywedge\omega)\geq k\}. Note that \{0\}=\dots={\mathcal{R}}_{-1}(W,\Bbbk)={\mathcal{R}}_{0}(W,\Bbbk)\subseteq{\mathcal{R}}_{1}(W,\Bbbk)\subseteq\dots\subseteq{\mathcal{R}}_{r}(W,\Bbbk)={\mathcal{R}}_{r+1}(W,\Bbbk)=\dots=A^{1}_{W,\Bbbk}, where r=\dim_{\Bbbk}A^{1}_{W,\Bbbk} . Both families of complexes are neither isomorphic nor quasi-isomorphic in general; however, one of our main results is Theorem 7 which states that there is a morphism of complexes that allows one to identify their resonance varieties, namely, {\mathcal{R}}_{k}(W,\Bbbk)\cong{\mathcal{R}}_{k}(X,\Bbbk). Combining this with [ 20 , Theorem C] we compute in § 4 , Theorem 11 , upper bounds on b_{k} for k a prime power, using combinatorial resonance varieties of the weak combinatorial type of the curve. We illustrate this result by computing bounds for the multiplicities of roots of the classical Alexander polynomials of order powers of primes, for a series of examples. We also present in § 4.1 a triviality criterion for the cohomology in degree 1 of the Combinatorial Aomoto complex with coefficients in a finite field. This in turn enables us to formulate combinatorial conditions to ensure b_{k}=0 , see Theorem 22 . Section § 4.2 is devoted to the twisted counterpart of the previous results. In addition, in § 4.3 we present some examples. A tool for obtaining lower bounds for b_{k} , and more generally for b_{k,\bar{m}} , is presented in § 5 and applied to (quasi) fiber-type curves , that is, the class of curves that can be obtained as a union of fibers of a pencil. These lower bounds depend on certain properties of the pencil. This class contains as a subclass the fiber–type curves, also known as pencil–type curves, in particular the Halpen pencils of index k\geq 1 defined in [ 10 ] , and the projective line arrangements that admit combinatorial pencil/multinet structures as defined in [ 5 , 13 ] . The lower bounds for b_{k,\bar{m}} in the (quasi) fiber-type curve case are given precisely by the number and multiplicity of the non-reduced fibers of the pencil \mathcal{C} is a union of, but also, and more surprisingly, by the extra multiple fibers that are not part of \mathcal{C} , see Theorem 37 . This generalizes several recent results, for instance [ 9 , Theorems 3.2 – 3.4] . It is worth noticing that this result applies not only to roots of unity of order prime powers as the results in section § 4 did. In § 6 we present a series of applications to reprove known results by Dimca-Pokora-Sticlaru in [ 9 ] on Halphen curves and to provide new examples of calculations of b_{k} when it coincides with the upper and lower bounds obtained above. Finally, in § 7 we revisit Yoshinaga’s