Has there been a paper on why GDI are causing fuel dilution? Given that the fuel is injected near the top of the compression stroke - and maybe in two shots for a stratified charge. I don't understand the mechanism for sump dilution over a PI fueler.
Does the gas wash on the cylinder walls in a PI engine help create a ring seal that is now missing on the GDI? It's got to be more than just injection pressure issue - the atomization must be better in GDI and the particle KE has to be less given the lower mass, so I don't think it's a "pressure wash past the rings" scenario.
With a PI engine the fuel has vaporised inside the manifold, before entering the compression chamber. Those engines tended to wet the walls of the intake manifold.
DI engines spray liquid fuel into the compression chamber. If the fuel doesn't vaporise quickly enough, there's liquid fuel reaching the cilinder walls eventually. That mixes with the oil present there, and won't heat up much beyond the cilinder wall temperature. That temperature isn't much higher than the coolant temp.