I really doubt this would work. The reason hardly .1% of academic work makes it into technology is because the vast majority of it has no practical application

I really doubt this would work. The reason hardly .1% of academic work makes it into technology is because the vast majority of it has no practical application whatsoever. And aside from that, there's a huge problem with the volume of material published. Even academics themselves have trouble keeping up with all the work that's published in their own narrow field. You can spend every day reading literature for years without once encountering a marketable idea.