Sure - the issue is that prefiltering is done based on the numbers. So if you've filtered out all of the good candidates, because all their publications are real, and one is left with a field full of candidates with CVs stuffed with junk publications, then the choices are a) mark them all unemployable, or b) hire one of the few mediocre candidates who may have some good publications amongst the myriad of junk ones.
This problem means that even the good candidates are forced to stuff their CVs with junk publications, just in order to get through the first round of culling.
I'm thankful I never had a career in academia - I'm proud of my publication record, all publications have a definite reason for existing, with little in the way of duplication or 'salami slicing', even though it looks less than stellar compared to my colleagues at a similar stage in their careers.