Just thought I'd share something I'm finding very useful. I've decided I want to read (and highlight) papers digitally since I'm otherwise perpetually printing out papers and losing them and having to read them again... Anyway! I've found that any papers before about 2005 rarely have the text built in to the PDF, which makes it hard to highlight. However you can use OCR to add text to a PDF - I've found this one works quite well:
You just upload the PDF, set the output format to PDF and it will add text to the document. So far it's done well, doesn't handle tables or images quite as well but it's good enough for me! Only downside is that it makes the resulting PDF quite large (a factor of 10 maybe), so beware of large file sizes. If you find a workaround post here!