That's a different "smart" feature -- the one mentioned has to do with guessing where the selected text continues (when it goes on to a different column, text box, etc).
PDFs are more "baked" layouts rather than high level semantic markup, so it's not easy to automatically determine this. The AI is probably a set of heuristics.
PDFs are more "baked" layouts rather than high level semantic markup, so it's not easy to automatically determine this. The AI is probably a set of heuristics.