[chagen]

What LLMs can do efficiently is crawl through and identify the secondary forms of evidence you mentioned. The real power behind retrieval architectures with LLMs is not the summarization part- the power comes from automating the retrieval of relevant documents from arbitrarily large corpuses which weren't included in the training set.

[asdff]

What makes a document relevant or not? Provenance? Certain keywords? A lot of this retrieval people cite that llms are good at can be done with existing search algorithms too. These are imo nicer because they will at least provide a score for the fit of the given document to the term.