Two-sphere partition functions and Gromov-Witten invariants (Q2439688)

An \({\mathcal N}=(2,2)\) supersymmetric, two-dimensional nonlinear sigma model with target space a Calabi-Yau threefold \(X\) (physically) flows to an \({\mathcal N}=(2,2)\) superconformal field theory in the infrared. Along the way, perturbative and non-perturbative corrections ``quantum'' deform the complexified Kähler moduli space of \(X\). The quantum Kähler moduli space is a (local, or projective) special Kähler manifold, governed by the Kähler potential \(K_{\text{Kähler}}(X)\) together with an integral structure on the tangent space. The Kähler potential can equivalently be replaced by the ``prepotential'' \(F_{\text{Kähler}}(X)\), which is a section of the square of the Hodge bundle. Either of these is determined by, and encodes, the quantum cohomology of \(X\), i.e., the deformation of the cohomology ring of \(X\) by genus \(0\) Gromov-Witten invariants. Traditionally, one predicts Gromov-Witten invariants of \(X\) by finding a ``mirror manifold'' \(Y\), which is a family of Calabi-Yau threefolds around a point of maximal degeneration, and identifying \(F_{\text{Kähler}}(X)=F_{\text{complex}}(Y)\). Certain Calabi-Yau manifolds admit an alternative (physical) description as an \({\mathcal N}=(2,2)\) supersymmetric gauge theory (a.k.a.\ GLSM). The class of manifolds with such a description is larger than that with known mirror manifolds. Recently, two groups of researchers (Benini et al., and Doroud et al.) have calculated the (physical) partition function \(Z_{S^2}(\text{GLSM)}\) of such a supersymmetric gauge theory on the two-sphere. The input is the gauge and matter content (i.e., a compact Lie group and a unitary representation), and certain couplings (Fayet-Iliopoulos parameters and a superpotential). The result takes the form of an infinite sum (over flux sectors), and a finite-dimensional integral (over the maximal torus of the gauge group), of a relatively simple rational Gamma function. It this paper, it is conjectured that \[ Z_{S^2}(\text{GLSM}) = \exp(-K_{\text{Kähler}}(X)) \] The main evidence is an evaluation of the left hand side, first for two examples with a known mirror, in which existing results are reproduced, and then for a new example (a determinantal Calabi-Yau variety introduced by Gulliksen-Negård), which leads to new predictions of Gromov-Witten invariants, some of which can be independently verified. The main formula can also be interpreted in terms of topological-anti-topological fusion (\(tt^*\)-geometry), which constitutes additional evidence for the proposal. The results of this paper have opened a very interesting window onto ``mirror symmetry without mirror symmetry'', and already generated a number of susequent works.