> I suspect that LLMs are better at classifying novel vs junk papers than they are at creating novel papers themselves.

Doubt

LLMs are experts in generating junk. And generally terrible at anything novel. Classifying novel vs junk is a much harder problem.