[nl]

I've done some work with graph neural nets as well as text NNs.

I think we've repeatably seen that models which replace an end-to-end system with a single model work amazingly well when there is sufficient data to train the whole system.

But there are often practical reasons why a non-end-to-end system are easier to build as an intermediate step.

[fho]

And, in theory, there is nothing stopping you from setting up a graph based system consisting of several small models and train that end-to-end.