Actually, I have tried it, as an experiment on a dark brew that originally somehow finished a bit too dry for my tastes (despite following a recipe I'd been using for years). I already drank through most of the keg unadulterated, but one day while making a batch of iced tea (stevia is fantastic in iced tea...and other things for that matter) it occurred to me to try it in the too-dry beer. I poured a half-pint and added about 1/8th of a packet of stevia to the glass...and it was actually pretty good.
And that 1/8th of a packet was just enough to make a difference. For what it's worth, there was no evidence of unpleasant aftertaste either. There was no hint of anything artificial in the taste of the beer.
So based on my experience with it, I say there's no reason at all to avoid experimenting with it to backsweeten in an emergency. I imagine that it would yield better results is some beers over others. But I think it's worth at least trying to save a disappointing brew until you brew the next batch with corrections.