I am a PhD Student works in computer science. I am currently in the middle stage of my PhD. I am facing a serious problem right now. It is about reading a research paper at a faster rate. I usually takes 2 weeks to read a paper nowadays, but my supervisor wants me to read a research paper in 4-5 days with greater detail. It involves mathematics and it takes time to get used to it. I have tried to explain him the problem that it is not possible at this time for me to read a paper in 4-5 days but keep pushing me to do so.

Question : How to read a research paper at faster rate?

  • Similar question : academia.stackexchange.com/q/95310/72855 – Solar Mike Jun 23 '18 at 17:09
  • 7
    What do you exactly mean by read a paper? Two extremes are: 1) read it in its entirety, understanding every single word/formula, and being able to reproduce its results in the smallest details; 2) read the abstract, introduction, discussion & conslusions to just know what&how was done - arrive at conclusions such as "ah, so they used X to verify if Y, and they found that Z; cool", then save the paper for possible future use. I've done 1) only a few times and it took up to weeks-months. I do 2) with several papers a day. – user68958 Jun 23 '18 at 17:11
  • 1
    It comes with time an experience. – Oleg Lobachev Jun 23 '18 at 17:18
  • 1
    @old It is impossible. – Andrés E. Caicedo Jun 23 '18 at 18:44
  • 4
    It sounds like you have an XY problem here: you think your problem is with your comprehension speed, but it sounds to me like your actual problem is in managing your relationship with your advisor (especially his expectations). – zibadawa timmy Jun 23 '18 at 19:21

Reading and understanding research papers in depth takes time! Two weeks sounds perfectly normal to me. I often need more time for indept understanding.

My strategy is that I read the core papers required for my research in great detail and other papers (context related) more generally. However, if I read quickly, I also tend to forget quickly.

Probably your supervisor is refering to his own (past?) experience or compares your speed with other researchers. However, this may not be a fair comparison. Learning and research styles are highly individual.

I notice that each scientist develops a certain strategy for reading papers. Ocassionally, I also see styles that I do not like. For example a scientist supporting a claim with reference to a source (of many pages) which didn’t even mentionned the subject although the title of the source suggested otherwise. The source clearly has not been read.

To conclude, you have to develop your own style. Ask your supervisor why he/she thinks you should be faster and try to solve the underlying sentiments. Maybe your supervisor is under pressure to deliver results.


You said it yourself,

It involves mathematics and it takes time to get used to it.

Thus, the solution is to get used to it. It took me a few months to read the first paper I read. Now I can skim through a few papers for breakfast and read one in a couple of hours. It just takes a lot of practice. Also, the more you learn, the less new things will be in each paper. Usually, with mathematics in particular, some ideas and details take a sentence to explain and understand, but takes several pages to explain in a rigorous manner. Learning to identifying the ideas and the main theme is what you seek, and then get down to the details only when you need to implement something.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.