One of the continuing struggles we have in representation is having to repeat the same tasks over and over without being able to store our reasoning process and repeat it in ways that will give us solutions with similar levels of rigor and precision each time.
A reasoning process that had its own form of representation that it could apply reliably to a class of problems would potentially save itself a great deal of repeated computation. If we could embed parts of our reasoning into a graph and embody them through compilation, we would save our reasoning and memory functionality a great deal of work.