Based morphisms for characters of quantum symmetric pairs
Stein Meereboer · 2025
Read Based morphisms for characters of quantum symmetric pairs by Stein Meereboer (2025) 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-2510-17655.txt.
Introduction
Stein Meereboer classifies special modules in quantum algebra in a 2025 arXiv paper. His objects are quantum symmetric pairs, algebraic structures used in modern representation theory. He studies their one dimensional modules, which are the simplest building blocks. The question is which of them appear inside simple finite dimensional modules of the larger algebra.
Meereboer gives a complete classification of those one dimensional modules. He also determines the branching rules, which say how big modules split when viewed through the smaller structure. The main result shows certain projections respect the based structure on both sides. To prove it, he builds a crystal basis theory at a limiting value. The results match the integral forms of canonical bases.
Catalog record
| CREATOR | Stein Meereboer |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 59 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2510-17655
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Based morphisms for characters of quantum symmetric pairs Stein Meereboer Abstract . We study based one-dimensional modules of quantum symmetric pairs over the field ℚ ( q ) \mathbb{Q}(q) . We provide a complete classification of one-dimensional 𝐁 \mathbf{B} -modules that appear as submodules of simple finite-dimensional based 𝐔 \mathbf{U} -modules and determine the corresponding branching rules. The main result of this paper shows that the corresponding projections are morphisms of based 𝐁 \mathbf{B} -modules. To this end we characterize one-dimensional modules at q = ∞ q=\infty , thus developing a ı \imath crystal basis theory for these modules. This is then applied to show compatibility with the integral forms of the (dual-)canonical basis. 2020 Mathematics Subject Classification Primary 17B37 1. Introduction 1.1. Quantum symmetric pairs Let \mathfrak{g} be a semisimple Lie algebra over \mathbb{C} and \Theta:\mathfrak{g}\to\mathfrak{g} an involutive Lie-algebra automorphism. The fixed point subalgebra of \Theta is denoted by \mathfrak{k}=\mathfrak{g}^{\Theta} . The pairs (\mathfrak{g},\mathfrak{k}) are known as symmetric pairs, in this paper we restrict our attention to the symmetric pairs of Hermitian type. The theory of quantum symmetric pairs, as developed by Letzter, provides quantum analogs of the pairs of universal enveloping algebras (U(\mathfrak{g}),U(\mathfrak{k})) [ 12 ] . Let (\mathbf{U},\mathbf{B}) be a quantum symmetric pair over the field \mathbb{Q}(q) , cf. [ 10 ] [ 12 ] . This pair consists of a Drinfel’d Jimbo quantum group \mathbf{U} and coideal subalgebra \mathbf{B}\subseteq\mathbf{U} whose structure is determined by a family of scalars (\bm{c},\bm{s}) . 1.2. Based \mathbf{B} -morphisms for the trivial module The theory of based \mathbf{B} -modules provides generalizations of Lusztig’s theory of based \mathbf{U} -modules [ 5 , Definition 6.11] . The existence of the \imath canonical basis [ 4 ] implies that every finite-dimensional based \mathbf{U} -module is naturally equipped with the structure of a based \mathbf{B} -module. In [ 18 ] , Watanabe proved the stability conjecture of \imath canonical bases [ 4 , Remark 6.18] . Watanabe’s proof of this conjecture is a consequence of [ 18 , Theorem 4.3.1] , which verified that distinguished \mathbf{B} -homomorphisms to the trivial module are morphisms of based \mathbf{B} -modules. Denote \breve{X}^{+} the subset of X^{+} consisting of spherical weights, see Section 3.2 . Theorem 1 . [ 18 , Theorem 4.3.1] Let \lambda\in\breve{X}^{+} . Then there exists a unique morphism as based \mathbf{B} -modules L(\lambda)\to L(0),\qquad v_{\lambda}\mapsto 1. Theorem 1 also formed the foundation for [ 3 ] , where it was applied to construct integral forms of symmetric spaces over algebraically closed fields of characteristic not equal to 2. 1.3. Results The main purpose of this paper is to generalize Theorem 1 to nontrivial one-dimensional modules of \mathbf{B} . Here, we remind the reader that the quantum symmetric pair (\mathbf{U},\mathbf{B}) is assumed to be a quantization of a symmetric pair of Hermitian type. In [ 15 ] , we classified the one-dimensional \mathbf{B} -modules appearing as submodules of finite-dimensional weight \mathbf{U} -modules and determined their branching rules. However, the field of definition is the bigger field k=\overline{\mathbb{C}(q)} , the algebraic closure of \mathbb{C}(q) . This was needed to import specialization results of [ 11 ] . It is natural to ask whether the given classification persists over the smaller field \mathbb{Q}(q)\subseteq k . For suitable parameters (\bm{c},\bm{s}) of the coideal subalgebra \mathbf{B} we give an affirmative answer. Theorem 2 (Proposition 5.1 ) . Let l\in\mathbb{Z} . Then there exists a character \chi_{l}:\mathbf{B}\to\mathbb{Q}(q) and a unique weight \mu_{l}\in X^{+} such that [L(\lambda)|_{\mathbf{B}}:\chi_{l}]=\begin{cases}1&\text{if\,\,}\lambda=\mu_{l}+\mu\text{ with }\mu\in\breve{X}^{+}\\ 0&\text{else}\end{cases}. Moreover, the characters \chi_{l} are pairwise inequivalent and provide a classification of the one-dimensional \mathbf{B} -modules that appear as submodules of finite-dimensional weight \mathbf{U} -modules. Consequently, for each l\in\mathbb{Z} , we obtain a one-dimensional module V_{\chi_{l}}=\mathrm{span}_{\mathbb{Q}(q)}\{1\} satisfying b1=\chi_{l}(b)1,\qquad\text{where}\qquad b\in\mathbf{B}. Theorem 3 (Theorem 6.3 ) . Let \lambda\in\breve{X}^{+} and let l\in\mathbb{Z} . Then there exists a unique morphism as based \mathbf{B} -modules L(\mu_{l}+\lambda)\to V_{\chi_{l}},\qquad\qquad v_{\mu_{l}+\lambda}\mapsto 1. This generalizes [ 18 , Theorem 4.3.1] to non-trivial one-dimensional modules. In order to prove Theorem 3 we follow the strategy outlined in [ 18 ] and [ 19 ] . As in [ 19 ] , we need to understand the one-dimensional modules at q=\infty . We characterize the \mathbf{B} -modules V_{\chi_{l}} in each highest weight \mathbf{U} -module L(\lambda) in terms of its (ordinary) crystal base. Following the approach in [ 18 ] , we reduce the general case to the rank one case. In the rank one case, we apply Theorem 1 to further reduce the problem to characterizing the one-dimensional modules V_{\chi_{l}} at q=\infty in its minimal realization L(\mu_{l}) . This final step is done by direct computation. The most important part in the final step is that non-trivial characters appear only in limited situations; there are just two cases to consider. Although in these cases there are infinitely many inequivalent characters, their behavior at q=\infty can be deduced from the minimal cases \chi_{1} and \chi_{-1} . Therefore, we only need to perform a small amount of case-by-case computation to adopt results from Watanabe. We write v=v_{\lambda}+\mathrm{l.o.t.} in the highest weight module L(\lambda) to indicate that \mathrm{l.o.t.}=\sum_{\mu<\lambda}v_{\mu} for weight vectors v_{\mu}\in L(\lambda)_{\mu} . Let \mathcal{A}=\mathbb{Z}[q,q^{-1}] and denote {}_{\mathcal{A}}L(\lambda) the integral form of L(\lambda) arising from the integral form of \mathbf{U} . Let \lambda\in\breve{X}^{+} and l\in\mathbb{Z} . By Theorem 3 there exists a vector f^{\lambda}_{l}=v_{\lambda}+\mathrm{l.o.t.}\in L(\mu_{l}+\lambda) spanning a module of type \chi_{l} . As an application of Theorem 6.3 we show that the vectors f^{\lambda}_{l} are compatible with integral forms. Theorem 4 (Theorem 7.4 ) . Let l\in\mathbb{Z} . Then f^{0}_{l}=v_{\lambda}+\mathrm{l.o.t}\in{{}_{\mathcal{A}}L(\mu_{l})} The proof of this result is based on the following remark: If \pi:{{}_{\mathcal{A}}}L(\lambda)\to{{}_{\mathcal{A}}V_{\chi_{l}}} is an \mathcal{A} -linear map then \pi^{\ast}({{}_{\mathcal{A}}V_{\chi_{l}}}^{\ast})\subseteq{{}_{\mathcal{A}}L}(\lambda)^{\ast} . In the literature the \mathcal{A} -module {{}_{\mathcal{A}}L}(\lambda)^{\ast} is known as the integral form of the dual canonical basis. We can identify L(\lambda) with its dual using a bilinear form. We denote the identification of {{}_{\mathcal{A}}L}(\lambda)^{\ast} by {}_{\mathcal{A}}^{\mathrm{up}}L(\lambda) , cf. [ 8 , (4.2.6)] . This identification often leads to a strict inclusion {}_{\mathcal{A}}L(\lambda)\subsetneq{{}_{\mathcal{A}}^{\mathrm{up}}L(\lambda)} . However, when the weight \lambda is (quasi-)minuscule, we have an equality (up to the zero weight space), cf. [ 7 , §9.21] . This equality allows us to identify the minimal spherical vectors in the integral form, which can be extended to all characters using tensor products. Acknowledgment The paper was inspired by conversations with Huanchen Bao during my visit to the NUS. I would to like to Huanchen Bao for the discussions and the hospitality. Moreover, I would like to thank Erik Koelink for corrections and valuable comments. This work is funded by grant number OCENW.M20.108 of the Dutch Research Council. 2. Quantum groups and and based modules 2.1. Quantum groups We introduce notation of quantum groups as in [ 13 ] . Let \mathfrak{g} be a complex semisimple Lie algebra with Cartan matrix C=(c_{ij})_{i,j\in I} . Let D=\operatorname{diag}(d_{i}\,:\,d_{i}\in\mathbb{Z}_{\geq 1},\,i\in I) be a symmetrizer, meaning that DC is symmetric and \gcd\{d_{i}\,:\,i\in I\}=1 . We fix a set of simple roots \Pi=\{\alpha_{i}\,:\,i\in I\} and a set of simple coroots \Pi^{\vee}=\{\alpha^{\vee}_{i}\,:\,i\in I\} with corresponding root and coroot systems \mathcal{R} and \mathcal{R}^{\vee} . The root lattice is denoted by \mathbb{Z}I=\oplus_{i\in I}\mathbb{Z}\alpha_{i} . The root lattice is equipped with the normalized Killing form (\,,\,) , scaled so that short roots have