Title: On the origin of long-range azimuthal correlations in hadronic collisions

Abstract: We review the models suggested, to date, as an explanation for the so called "ridge" phenomenon, an elongation in rapidity of 2-particle correlations seen at RHIC and LHC energies. We argue that these models can be divided into two phenomenologically distinct classes: "Hotspot+flow" driven correlations, where initial state correlations created by structures local in configuration space are collimated by transverse flow, and models where the azimuthal correlation is created through local partonic interactions in a high gluon density initial state.
We further argue that the key to distinguish these two scenarios is to experimentally examine the role flow plays in the ridge. The measurement of a strong double ridge in $pA$ and $dA$ collisions allows a good opportunity to understand the ridge's origin. In this sense, particle-identified correlations are particularly promising way of testing the assumptions at the base of either of the two models above.
The successful fit of the pA ridge by hydrodynamic models \cite{us}, therefore, suggests that the ridge is primarily a "final state effect" generated by radial expansion, although its scaling across system sizes is non-trivial to reconcile with hydrodynamic expectations. We close by discussing experimental observables capable of clarifying the situation.