In the autorotation if you push the cyclic forward the RPM will decrease.
I believe this is caused by a decrease in the angle of attack. There is now a less component force pulling the blade forward (driving region) and RPM will decrease. Conversely, aft cyclic will cause an increase in the angle of attack and an increase in the component force to pull the blade forward thereby increasing RPM.
Perhaps another way to look at, in an autorotation if you pitched down 90 degrees how much airflow would be coming up through the disk, what would the angle of attack be? I’m sure the RPM would drop way off in that situation.