Hacker News Reader

What determines relevance? It must do something other than page rank. Will it recognize synonyms and more subtle kinds of nearness in word-space?

Actually, the page rank is really based on semantic similarity and relevance of the matched paragraph.

Which under the hood is based of a t5 encoder

Does it work well for question answering from books ?

The tricky part is to understand which parts are you indexing, paragraph vs. sentences vs. pages.

But yeah