Reading Research Papers

Oct 11, 2015 - 2 mins

Lately, I’ve read my fair share of research papers. Along the way I’ve picked up a thing or two about how to read them. This is by no means a comprehensive best practice list of sorts; just a few tips I’d like to share that have worked for me.

  1. Don’t treat it like a novel. You don’t have to read this thing front to back. Jump back and forth where it makes sense. I tend to read the abstract and conclusion first, this gives me an idea of what the paper is about and the result of the findings. If this looks good I dig a bit deeper, usually jumping right to the contribution/experiments.

  2. Once you have your feet wet, skip the generic stuff. You’ll notice a decent chunk of any paper is going over the prior art. Often this is the same for many papers, once you know it, feel free to skip it.

  3. You don’t need to read them all. Reading a research paper is hard and requires brainpower, so be selective.

  4. Have purpose. Have a reason why you’re reading the paper, it’ll help you stick with it.

  5. Take notes. This is more about learning than anything else. It’s been shown taking notes while reading improves understanding significantly vs. just reading. I think this is even more true with research papers.

  6. Have fun. The stuff you’re reading is cool, so why not?