Quantum Riemann Surfaces in Chern-Simons Theory
Tudor Dimofte
TL;DR
The paper develops a principled, three-dimensional construction to quantize the A-polynomial of knot complements by recasting gluing in Chern-Simons theory as a symplectic reduction in a TQFT framework. It builds from single-tetrahedron quantization, using the noncompact quantum dilogarithm as the tetrahedral block, and glues many tetrahedra via a Weil-representation-based approach to produce a finite-dimensional state-integral model for the holomorphic blocks annihilated by the quantum A-polynomial \hat{A}. The authors demonstrate triangulation- and path-independence of the resulting operator and wavefunction data, and provide explicit quantum A-polynomials for several knots (e.g., $4_1$, $3_1$, $5_2$) along with their dual objects under modular S-duality. The framework links hyperbolic geometry, quantum topology, and the theory of holomorphic blocks, yielding both a constructive quantization of knot invariants and a practical state-integral method to compute wavefunctions and recursions. This work thus offers a robust, geometry-driven route to quantum A-polynomials with broad generalizations to links, higher-genus boundaries, and higher-rank groups, enriching the interface between quantum field theory and low-dimensional topology.
Abstract
We construct from first principles the operator 'A-hat' that annihilates the partition functions (or wavefunctions) of three-dimensional Chern-Simons theory with gauge groups SU(2), SL(2,R), or SL(2,C) on a knot complement M. The operator 'A-hat' is a quantization of the knot complement's classical A-polynomial A(l,m). The construction proceeds by decomposing three-manifolds into ideal tetrahedra, and invoking a new, more global understanding of gluing in TQFT to put them back together. We advocate in particular that, properly interpreted, "gluing = symplectic reduction." We also arrive at a new finite-dimensional state integral model for computing the analytically continued "holomorphic blocks" that compose any physical Chern-Simons partition function.
