Euler characteristics in the quantum \(K\)-theory of flag varieties (Q1987525)

The paper under review is concerned with the quantum \(T\)-equivariant \(K\)-theory of a homogeneous variety \(X:=G/P\), where \(G\) is a connected semisimple complex linear algebraic group, \(P\) a parabolic subgroup and \(T\) any maximal subtorus. Since \(X\) is obviously acted on by \(T\), it is natural to wonder about the \(T\)-equivariant Grothendieck ring \(K_T(X)\) of \(T\)-equivariant vector bundles on the homogeneous variety \(X\), which is an algebra over \(\Gamma:=K_T(pt)\), the equivariant Grotendieck ring of a point. The beautiful main result of the paper under review is the determination of the quantum equivariant product of the class of a Schubert variety with its \textit{opposite} in the Grothendieck ring of the generalised flag varieties. To be more precise, for the convenience of the potential reader's audience, recall that there are two ways to define Schubert varieties in \(X\) starting by elements of the Weyl group of \(G\). If \(B\) is a Borel subgroup such that \(T\subseteq B\subseteq P\), the opposite Borel subgroup \(B^{-}\) is characterised by the equality \(B\cap B^{-}=T\). Each element \(w\) of the Weyl group of \(G\) hence determines a \(B\)-invariant Schubert variety \(X_w\) and a \(B^{-}\)--invariant one \(X^w\), which are defined to be one the \textit{opposite} of the other. Another piece of information is that the second cohomology group \(H^2(X,Z)\) possesses a basis \([X_{s_\beta}]\), where \(s_\beta\) is the element of the Weil group of \(G\) corresponding to \(\beta\in \Delta\setminus \Delta_P\), where \(\Delta\) is the set of the simple positive roots of \(G\) and \(\Delta_P\) those of \(P\). Given a formal variable \(q\), one attaches a formal variable \(q^\beta\) to each simple root \(\beta\) and define \(\Gamma[[q]]\) as the ring of formal power series in the variables \(q^\beta\). Let \(q^d:=\prod_{\beta\in \Delta\setminus\Delta_P}q^{d_\beta}_\beta\). One defines, à la \textit{A. Givental} [Mich. Math. J. 48, 295--304 (2000; Zbl 1081.14523)], the small quantum Grothendieck ring \(QK_T(X)\) as being the tensor product of \(K_T(X)\) with \(\Gamma[[q]]\) with respect to a product which deforms the classical one via corrections coming from \(K\)-theoretical Gromov-Witen invariants. This is very well explained already in the article's introduction. The main character of the paper is a suitable quantum extension of the \textit{sheaf Euler characteristic} \(\chi_X:K_T(X)\rightarrow \Gamma\), which maps a class \([E]\) of equivariant vector bundles to its generalised Euler characteristic \(\chi_X([E])\), namely to the formal sum \(\sum_{i\geq 0}(-1)^i[H^i(X,E)]\). The latter can however be defined as the unique map such that \(\chi_X([\mathcal{O}_{X_u}])=\chi_X([\mathcal{O}_{X^v}])=1\), for all \(u,v\in W\). Consider the \(\Gamma[[q]]\)-linear extension \(\chi\) of \(\chi_X\). The central goal of the paper is then to prove that such sheaf Euler characteristic \(\chi\) of the product of a Schubert class and an opposite Schubert class in the quantum \(K\)-theory ring of a (generalized) flag variety \(G/P\) is equal to \(q^d\), where \(d\) is the smallest degree \(\sum d_\alpha X_{s_\alpha}\) of a rational curve joining the two Schubert varieties. This implies that the sum of the structure constants of any product of Schubert classes is equal to \(1\). Along the way, a description of the smallest degree \(d\) in terms of its projections to flag varieties defined by maximal parabolic subgroups is provided. This result was already proved by \textit{A. S. Buch} and \textit{S. Chung} [J. Lond. Math. Soc., II. Ser. 97, No. 2, 145--148 (2018; Zbl 1390.14169)] for cominuscule Flag varieties. The paper consists of just three sections. Section one is a very well written and explicative introduction: the reader gets immedately aware about what the authors are concerned with. Section 2 recall and explain the interesting notion of distance between two Schubert varieties and Section 3, finally, is devted to provide the last missing preliminaries in order to prove the main Theorem 8, we have mentioned above. The essential reference list conclude this very inspiring paper.