The map between large-N conformal field theories and semiclassical gravity has been one of the defining achievements of holography. However, the large N holographic dictionary remains incomplete. One of its most notable criticisms, is the failure to address the factorization problem, where the appearance of Euclidean wormholes in the gravitational path integral, lacks a clear interpretation on the large N CFT side. A related challenge is the possibility of erratic N dependence in CFT observables, behaviour with no evident semiclassical gravitational counterpart. In arXiv:2512.13807, a solution is proposed in the form of a large N filter that removes the erratic N dependence of CFT quantities and provides a boundary explanation of  wormhole contributions.
In this talk, I will briefly review the factorization problem and illustrate the proposed large N filter resolution. Time permitting, I will also outline some of the Lorentzian spacetime structures that can emerge when working within the framework of such a large N filter, such as the appearance of baby universes and black holes interiors.

Traditionally, a symmetry of a quantum system refers to a transformation that preserves transition probabilities between physical states. In recent years, this notion has been expanded to so-called generalised symmetries, which correspond to (possibly non-invertible) topological defects in quantum field theory. At first sight, it is not obvious how the above two notions of symmetry are related. In this talk, I will review the notion of generalised symmetries and discuss how they relate to (and depart from) the traditional notion of symmetry.