On solutions of the Bethe Ansatz for the Quantum KdV model

TL;DR

This paper proves the completeness of solutions to the Bethe Ansatz equations for the Quantum KdV model, establishing a one-to-one correspondence with monster potentials and confirming the ODE/IM conjecture for degrees greater than 2.

Contribution

It provides a full classification of solutions with positive roots, proves the ODE/IM correspondence, and introduces a new mathematical framework via nonlinear integral equations.

Findings

Complete classification of solutions with positive roots.

Proof of the one-to-one correspondence between solutions and monster potentials.

Validation of the ODE/IM correspondence for degree > 2.

Abstract

We study the Bethe Ansatz Equations for the Quantum KdV model, which are also known to be solved by the spectral determinants of a specific family of anharmonic oscillators called monster potentials (ODE/IM correspondence). These Bethe Ansatz Equations depend on two parameters, identified with the momentum and the degree at infinity of the anharmonic oscillators. We provide a complete classification of the solutions with only real and positive roots -- when the degree is greater than 2 -- in terms of admissible sequences of holes. In particular, we prove that admissible sequences of holes are naturally parameterised by integer partitions, and we prove that they are in one-to-one correspondence with solutions of the Bethe Ansatz Equations if the momentum is large enough. Consequently, we deduce that the monster potentials are complete, in the sense that every solution of the Bethe Ansatz Equations coincides with the spectrum of a unique monster potential. This essentially (i.e. up to gaps in the previous literature) proves the ODE/IM correspondence for the Quantum KdV model/monster potentials -- which was conjectured by Dorey-Tateo and Bazhanov-Lukyanov-Zamolodchikov -- when the degree is greater than 2. Our approach is based on the transformation of the Bethe Ansatz Equations into a free-boundary nonlinear integral equation -- akin to the equations known in the physics literature as DDV or KBP or NLIE -- of which we develop the mathematical theory from the beginning.