The Non-melonic Sector of Tensor Models and Gravity

Abstract: The melonic sector has been proven to be dominant in tensor models at large N. This is true as long as the observables we consider, composites of 2n tensors, are small. That is, if n is much smaller than N. In this paper, I argue that, in order to recover geometries (and then gravity) in the continuum limit, n must grow like N. In that case, I provide examples where non-melonic contributions overcome the total sum in the computation of the expectation value of certain observables.