Yes it’s just an engineering problem, there is no a priori reason not to do it.
Maybe my simple scheme would be all it needs. Or maybe it needs some new breakthrough and right now nobody knows how to do it. I was hoping some resident expert could let me know.
First, it seems possible that if sources were in the training data like I described, then understanding of sources could be an emergent capability, just because the LLM reads "the source of the following is X."
Second, maybe a trainer LLM could be tasked with reading the trainee's answers and any sources it provides, and judging whether the source is correct.
But I'm no expert, hence my question.
In operation, the "trainer" could do the same thing in the background. And then of course, human users could also check up on the sources if they need to be sure of catching hallucinations.