Counting points on braid varieties and the Deligne–Simpson problem
Masoud Kamgarpour, Bailey Whitbread · 2026
Read Counting points on braid varieties and the Deligne–Simpson problem by Masoud Kamgarpour, Bailey Whitbread (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-2603-20499.txt.
Introduction
Kamgarpour and Whitbread solve the isoclinic Deligne-Simpson problem for exceptional groups. This finishes a program started by earlier researchers. The problem asks when certain local systems with prescribed data exist. Their route uses the Riemann-Hilbert correspondence to change settings. Emptiness questions become point counts on braid varieties over finite fields. Those varieties attach to periodic braids in the problem. As a by-product, they gain new rigid irregular connections. The method echoes Lusztig's map between Weyl group and unipotent classes.
Catalog record
| CREATOR | Masoud Kamgarpour, Bailey Whitbread |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33150 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.20499
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Counting points on braid varieties and the Deligne–Simpson problem Masoud Kamgarpour, Bailey Whitbread Abstract . We solve the isoclinic Deligne–Simpson problem for exceptional groups, completing a program initiated by Sage et al. and Jakob–Yun. As a by-product, we obtain new examples of physically rigid irregular G G -connections on ℙ 1 \mathbb{P}^{1} . Our approach uses the Riemann–Hilbert correspondence to reduce the problem to determining the non-emptiness of certain braid varieties associated to periodic braids. We show that this can be achieved by counting points over finite fields. Our approach is inspired by Lusztig’s construction of a map from conjugacy classes in the Weyl group to unipotent classes. 1. Introduction Let G be a complex reductive group. The Deligne–Simpson problem concerns the existence of G -connections on curves with prescribed singularities at finitely many points. For G=\mathrm{GL}_{n} , substantial progress has been made using non-abelian Hodge theory [ Simpson ] , \ell -adic cohomology [ Katz ] , and quiver varieties [ CB , Hiroe ] . In contrast, much less is known for groups of general type. In this paper, we solve the Deligne–Simpson problem for a class of G -connections on \mathbb{P}^{1}-\{0,\infty\} known as isoclinic . 1 1 1 Our terminology is consistent with [ JY ] but different from [ SageReview ] . In particular, what we call isoclinic is called toral in [ SageReview ] . Following [ HJ ] , we reserve the term toral for a more general notion; see § 2 . Such connections have an irregular singularity at 0 with a regular semisimple Levelt–Turrittin leading term, and a regular singularity at \infty . Important examples include generalised Frenkel–Gross connections (also known as generalised Kloosterman connections) [ FrenkelGross , HeinlothNgoYun ] , Yun’s \theta -connections [ YunEpipelagic , ChenRigid ] , and Coxeter connections [ KS-rigid , KS-galois ] . Suppose G is a simple group and let W denote its Weyl group. Then it is known that the slope of an isoclinic connection is of the form \nu=d/m , where m is a regular number for W (in the sense of Springer [ Springer ] ) and d is a positive integer coprime to m , cf. [ JY , Lemma 2.2] . Our main result is as follows: Theorem 1 . There exists a (necessarily unique) unipotent conjugacy class \mathcal{C}_{\nu}\subset G such that there is an isoclinic G -connection of slope \nu and unipotent monodromy \mathcal{C} if and only if \mathcal{C}\geq\mathcal{C}_{\nu} . 2 2 2 Here \geq denotes the usual partial order on unipotent conjugacy classes: \mathcal{C}\geq\mathcal{C}^{\prime} if and only if \overline{\mathcal{C}}\supseteq\mathcal{C}^{\prime} . When \nu>1 , we have \mathcal{C}_{\nu}=\{1\} . When \nu<1 and G is of exceptional type, the classes \mathcal{C}_{\nu} are listed in § A.1 . When G=\mathrm{GL}_{n} and m=n , this result was established by Sage and collaborators [ SageEtAl , LivesaySageNguyen ] using intricate linear algebra. Jakob and Yun [ JY ] proved the theorem for classical groups via non-abelian Hodge theory. They also treated some exceptional cases; however, groups of type E remained unresolved. Our approach for exceptional groups is different and uses instead the Riemann–Hilbert correspondence. The version of the Riemann–Hilbert correspondence we need was established recently by Hohl and Jakob [ HJ ] , building on prior work of Boalch [ BoalchTopology ] . The key input is that irregular connections give rise to positive braids [ BBAMY ] . The moduli space of irregular connections is then identified with the corresponding braid variety. These varieties first appeared in the work of Deligne on the K(\pi,1) Conjecture [ Deligne ] , and have recently attracted considerable interest due to their central role in symplectic topology, knot invariants, and cluster algebras [ STZ , Trinh , GorskyKivinenSimental , IanEtAl , Mellit , GLSB ] . To take into account the monodromies of connections, we need to work with variants of braid varieties introduced by Trinh [ Trinh ] . It turns out that the class of braids arising from isoclinic connections coincide with periodic braids, i.e., those with a non-trivial central power. More precisely, we show that the braid associated to an isoclinic connection of slope \nu=d/m is (\widetilde{w_{m}})^{d} , where w_{m} is an m -Springer element (in the sense of [ BM ] ), and \widetilde{w_{m}} denotes its lift to the positive braid monoid. Deep results of Broué–Michel and Bessis [ BM , BessisKPi1 ] imply that, up to conjugacy in the braid group, \widetilde{w_{m}} is the unique m^{\text{th}} -root of the full twist. The above discussion implies that the existence of an isoclinic connection of slope \nu and monodromy \mathcal{C} is equivalent to the non-emptiness of the braid variety associated to (\widetilde{w_{m}})^{d} and \mathcal{C} . We shall see that this in turn is equivalent to the condition (Bw_{m}B)^{d}\cap\mathcal{C}\neq\emptyset. The problem of determining when the intersection of a Bruhat double coset with a conjugacy class is non-empty has been studied extensively and is known to be difficult even for \mathrm{GL}_{n} , cf. [ EG04 , EG07 , CLT10 , Lusztig , LusztigCSmall , Malten ] . When w is an element of minimal length in its conjugacy class, Lusztig proved that there exists a (necessarily unique) unipotent conjugacy class \mathcal{C}_{w} such that a unipotent class \mathcal{C} satisfies BwB\cap\mathcal{C}\neq\emptyset if and only if \mathcal{C}\geq\mathcal{C}_{w} . This yields Lusztig’s map \displaystyle\{\textrm{conjugacy classes in }W\} \displaystyle\longrightarrow\{\textrm{unipotent conjugacy classes in }G\} \displaystyle[w] \displaystyle\longmapsto\mathcal{C}_{w}. For groups of exceptional type, Lusztig’s proof relied on counting points over finite fields [ Lusztig ] . As he observed, the count reduces to character values of the finite reductive group G(\mathbb{F}_{q}) and the finite Hecke algebra H_{q} . Our proof of Theorem 1 for exceptional groups follows the same strategy; that is, we count points on the relevant braid varieties (more precisely, braid stacks) over finite fields. As in Lusztig’s case, the count reduces to character theory of G(\mathbb{F}_{q}) and H_{q} . The necessary character computations are carried out with the CHEVIE package for Julia developed by Michel [ Michel ] . Note that m -Springer elements and their powers do not usually have minimal length in their conjugacy classes. Moreover, in general, we need to take powers of Bruhat double cosets. Thus, our main theorem is not a formal consequence of Lusztig’s result. As a byproduct of Theorem 1 , we obtain new examples of cohomologically rigid isoclinic connections; see § A.2 . We show that these connections are physically rigid by proving that the corresponding braid varieties are singletons. Constructing the \ell -adic analogues of these connections and their Hecke eigensheaves is an interesting open problem. Finally, it is natural to ask for explicit realisations of the isoclinic connections whose existence is guaranteed by our main theorem. We make partial progress in this direction by establishing Conjecture 5.7.(2) of [ SageReview ] , concerning Coxeter G -connections with minimal monodromy. 1.1. Structure of the text In § 2 , we introduce toral connections (which are a generalisation of isoclinic connections), and discuss the Riemann–Hilbert correspondence and the Deligne–Simpson problem for them. We also recall Lusztig’s theorem and outline his proof. In § 3 , we focus on toral connections that are isoclinic. We first give an explicit description of their associated braids.