Eynard-Orantin B-model and its application in mirror symmetry (Q2313887)

The article under review is a survey about the Eynard-Orantin recursive algorithm on a spectral curve and its many spectacular applications in mirror symmetry and enumerative geometry. Mirror Symmetry is a deep relationship between a symplectic \(A\)-model geometry and its mirror algebraic \(B\)-model geometry. One way it manifests itself is by turning difficult enumerative problems on the \(A\)-side into tractable algorithmic methods on the \(B\)-side. When the \(B\)-model can be described as a spectral curve, then there is the powerful Eynard-Orantin topological recursion associated to the spectral curve that computes all genus \(A\)-model invariants of the \(A\)-side. One large class of such \(A\)-model geometries are semi-projective toric Calabi-Yau 3-orbifolds for which the BKMP (Bouchard-Klemm-Mariño-Pasquetti) remodeling conjecture predicts its all genus open and closed Gromov-Witten invariants from topological recursion on its mirror curve. This, including some cases where it is proven, is surveyed in Section 4. In Section 2, the author introduces topological recursion. Let \(\Sigma\) be a smooth affine algebraic curve in \((\mathbb{C}^*)^2\) and let \(\overline{\Sigma}\) be a compact Riemann surface that compactifies \(\Sigma\). A choice of Torelli marking on \(\overline{\Sigma}\) turns \(\mathrm{H}_1(\overline{\Sigma},\mathbb{C})\) into a symplectic vector space via the intersection pairing. It gives rise to a certain symmetric meromorphic form \(B\) on \((\overline{\Sigma})^2\) called the \textit{fundamental normalized differential of the second kind}. Let moreover \(x\) be a holomorphic Morse function (superpotential) from the universal cover of \(\Sigma\) to \(\mathbb{C}^*\) with the property that \(\mathrm{d } x\) descends to a meromorphic form on \(\overline{\Sigma}\) with poles in \(\overline{\Sigma}\setminus \Sigma\). The datum of \(\boldsymbol{\Sigma} = (\Sigma,x,B)\) then is called a \textit{spectral curve}. Then Eynard-Orantin topological recursion produces a series of symmetric meromorphic forms \(\omega_{g,n}\) on \((\Sigma)^n\) called the \textit{\(B\)-model higher-genus invariants}. Suitable operations on these invariants produce the \(A\)-model invariants. For some \(A\)-model geometries including toric manifolds and their genus-0 Gromov-Witten \(A\)-model invariants, the \(B\)-model can be described as a semi-simple Frobenius manifold. They admit Givental quantization of the Frobenius manifold, which corresponds to the higher-genus Gromov-Witten invariants on the \(A\)-side. Section 3 explains how that structure is equivalent to an analogous structure on spectral curves, which includes oscillatory integrals. The author strikes a good balance between content and detail, surveying a good number of structures and giving sufficient detail. The many examples are particularly appreciated. What is missing is a geometric explanation relating the \(A\)-model, the Frobenius manifolds, the spectral curves and the mirror maps. Nonetheless, given the rapid development of topological recursion, this survey should be welcome to the reader wishing to learn about the domain. For the entire collection see [Zbl 1411.14003].