The image of functor morphing
Ehud Meir · 2026
Read The image of functor morphing by Ehud Meir (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-2603-26368.txt.
Introduction
Meir studies which representations can arise through functor morphing. The method turns representations of automorphism groups of finite modules over finite rings into representations for groups tied to functors. He gives a clear criterion for when a representation lies in the image of this map. The test rests on how parabolic subgroups act. He then shows the criterion at work on Borel groups over finite fields, checking which of their representations the morphing map can reach.
Catalog record
| CREATOR | Ehud Meir |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 68910 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.26368
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
The image of functor morphing Ehud Meir Abstract . Functor morphing provides a method to translate complex representations of automorphism groups of finite modules over finite rings to representations of automorphism groups of functors in some abelian category. In this paper we give an explicit criterion for a representation to be in the image of functor morphing using the action of parabolic subgroups. We then demonstrate this criterion on Borel groups of finite fields. 1. Introduction Let M be a finite module over a finite ring R . In [ CMO4 ] Tyrone Crisp, Uri Onn, and the author, introduced a new technique called functor morphing to study the representation theory of \Aut_{R}(M) . Assume that M=M_{1}^{a_{1}}\oplus\cdots\oplus M_{n}^{a_{n}} is the decomposition of M into a direct sum of non-isomorphic indecomposable modules. Functor morphing associates to every irreducible representation V of \Aut_{R}(M) an additive functor F:\langle M_{1},\ldots,M_{n}\rangle\to Ab and an irreducible representation \widetilde{V} of \Aut(F) . Here \langle M_{1},\ldots,M_{n}\rangle stands for the full additive subcategory of R-\,\mathrm{mod}\, generated by M_{1},\ldots,M_{n} . In most cases the procedure of functor morphing reduces the numerical complexity of the representation V . The functor F is the minimal (with respect to the subquotient relation) functor that satisfies \Hom_{\Aut_{R}(M)}(KF(M),V)\neq 0 , and the representation \widetilde{V} is given by \Hom_{\Aut_{R}(M)}(KF(M),V) , where \Aut(F) acts on KF(M) in the natural way. The fact that F is uniquely defined and that \widetilde{V} is irreducible follows from a construction of the author of symmetric monoidal categories using invariant theory [ M1 ] . The study of the representation theory of the groups \Aut_{R}(M) stems from the representation theory of the groups \GL_{n}(\mathfrak{o}/(\pi^{l})) , where \mathfrak{o} is a discrete valuation ring with uniformizer \pi and finite quotient field. One of the main philosophical consequences of functor morphing is the fact that the representation theory of these group contains in fact the representation theory of \Aut_{R}(M) for very general R and M . See also [ CMO1 , CMO2 , CMO3 ] for a study of the representation theory of \GL_{n}(\mathfrak{o}/(\pi^{l})) using generalizations of Harish-Chandra functors and Clifford Theory. Functor morphing thus gives a new way to study the representation theory of a very large class of finite groups. We get a stratification \Irr(\Aut_{R}(M))\cong\sqcup_{F}\Irr^{M,fm}(\Aut(F)) , where \Irr^{M,fm}(\Aut(F)) stands for the irreducible representations of \Aut(F) that arise from functor morphing of representations of \Aut_{R}(M) . This raises the question of what irreducible representations actually appear in \Irr^{M,fm}(\Aut(F)) . The goal of the present paper is to answer this question, and provide concrete examples for the family of upper triangular groups B_{n}(\mathbb{F}_{q}) . To state our main result, we introduce some terminology. Let M=\bigoplus_{i}M_{i}^{a_{i}} , F , and \mathcal{C}=\langle M_{1},\ldots,M_{n}\rangle , be as above. We say that (X\mathrel{{\mathop{\to}\limits}}Y) is a minimal presentation of F if \Hom_{R}(Y,-)\mathrel{{\mathop{\to}\limits}}\Hom_{R}(X,-)\to F\to 0 is a minimal resolution of F in the category \mathcal{C}_{+1}=Fun(\mathcal{C},\text{Ab}) . We have already seen in [ CMO4 ] the connection between the minimal resolution of F and functor morphing. In Section 3 we prove the following: Lemma 1.1 (See Lemma 3.18 ) . A necessary condition for representations of \text{Aut}_{\mathcal{C}_{+1}}(F) to appear as functor morphing of representations of \Aut_{R}(M) is that if (X\mathrel{{\mathop{\to}\limits}}Y) is a minimal presentation of F then X is a direct summand of M . If G\subseteq F is a subfunctor, we write Z_{G} for the unique object in \mathcal{C} such that \Hom_{R}(Z_{G},-) is the projective cover of G . We write P_{G}\subseteq\text{Aut}_{\mathcal{C}_{+1}}(F) for the subgroup \{\varphi\in\text{Aut}_{\mathcal{C}_{+1}}(F)|\text{Im}(\varphi-\text{Id}_{F})\subseteq G\} . In Section 4 we prove our main result: Theorem 1.2 . Let M , \mathcal{C} , and F be as above. Assume that F has a minimal presentation (X\mathrel{{\mathop{\to}\limits}}Y) and that M=X\oplus Z . Let V be an irreducible representation of \text{Aut}_{\mathcal{C}_{+1}}(F) . Then V is the functor morphing of an irreducible representation of \Aut_{R}(M) if and only if the following condition holds: for every semisimple subfunctor G\subseteq F such that Z_{G} is not a direct summand of Z , it holds that \Hom_{P_{G}}(\mathbf{1},V)=0 . The main result has the following corollary, characterizing the situations in which all the irreducible representations of \text{Aut}_{\mathcal{C}_{+1}}(F) arise from functor morphing: Corollary 1.3 . Let M=X\oplus Z and let (X\xrightarrow{f}Y) be a minimal resolution for F . Then the following conditions are equivalent: (1) Every irreducible representation of \text{Aut}_{\mathcal{C}_{+1}}(F) is the functor morphing of an irreducible representation of \Aut_{R}(M) . (2) The trivial representation \mathbf{1} of \text{Aut}_{\mathcal{C}_{+1}}(F) is the functor morphing of some irreducible representation of \Aut_{R}(M) . (3) If \Hom(X_{1},-) is the projective cover of soc(F) then X_{1} is a direct summand of Z . Finally, in Section 5 we give examples. We first consider the family of groups \GL_{n}(\mathbb{F}_{q}) . By [ Z1 ] we know that these groups give rise to a positive self-adjoint Hopf algebra \mathcal{H} . We explain the connection between our main result and the structure of \mathcal{H} . The second family of examples that we give is of the family of groups B_{n}(\mathbb{F}_{q}) of upper triangular matrices over \mathbb{F}_{q} . We show how our result translates to a combinatorial criterion, and we show how this gives a new enumeration of the irreducible representations of B_{n}(\mathbb{F}_{q}) when n\leq 4 . A long standing conjecture of Higman from 1960 [ H1 ] states that for every n the number of conjugacy classes in the unipotent radical U_{n}(\mathbb{F}_{q}) of B_{n}(\mathbb{F}_{q}) is a polynomial in q [ VA1 , So1 ] . Since the number of conjugacy classes is the same as the number of irreducible representations, and since U_{n}(\mathbb{F}_{q}) is a normal subgroup of B_{n}(\mathbb{F}_{q}) with an abelian quotient, this suggests that there are deep connections between enumeration of irreducible representations of B_{n}(\mathbb{F}_{q}) and Higman’s conjecture. It is our hope that, following this paper, the technique of functor morphing will be further used to study this conjecture. 2. Preliminaries 2.1. Functor morphing Let R be a finite ring and let M be an R -module. Functor morphing is a tool that was introduced in [ CMO4 ] to study the representation theory of \Aut_{R}(M) . We recall here the details. Write M=M_{1}^{a_{1}}\oplus\cdots\oplus M_{n}^{a_{n}} for the decomposition of M into non-isomorphic indecomposable modules. Write \mathcal{C}=\langle M_{1},\ldots M_{n}\rangle\subseteq R-\,\mathrm{mod}\, for the full subcategory of finite R -modules of the form \bigoplus_{i=1}^{n}M_{i}^{b_{i}} and write \mathcal{C}_{+1}=Fun(\mathcal{C},Ab) for the category of additive functors from \mathcal{C} to the category of finite abelian groups. The isomorphism classes of the objects in \mathcal{C}_{+1} are partially ordered, where F\preccurlyeq G if and only if F is isomorphic to a subquotient of G . For every F\in\mathcal{C}_{+1} it holds that F(M) is an \text{Aut}_{\mathcal{C}_{+1}}(F)\times\Aut_{R}(M) -set in a natural way. As a result, the linearlization KF(M)=span\{u_{v}\}_{v\in F(M)} is an \text{Aut}_{\mathcal{C}_{+1}}(F)\times\Aut_{R}(M) -representation. Theorem 2.1 (Theorem A, [ CMO4 ] ) . Let V be an irreducible representation of Aut_{R}(M) . There is a unique (up to isomorphism) functor F\in\mathcal{C}_{+1} such that \Hom_{Aut_{R}(M)}(KF(M),V)\neq 0 and \Hom_{Aut_{R}(M)}(KG(M),V)=0 for every G\prec F . This theorem shows that to every irreducible representation V of \Aut_{R}(M) we can attach a functor F\in\mathcal{C}_{+1} . In fact, more is true. Write \overline{KF(M)} for the largest subobject of KF(M) that satisfies \Hom_{\Aut_{R}(M)}(\overline{KF(M)},KG(M))=0 for every G\prec F . The \text{Aut}_{\mathcal{C}_{+1}}(F) -action on KF(M) restricts to an action on \overline{KF(M)} and we have an algebra