notes for an Artem Kirsanov video
Research Papers, while a very useful tool for getting practical experience with the academic world, can be somewhat dense and difficult to approach, especially if you are a undergraduate like me. This video outlined some useful steps to make sure you are getting the most effective use out your time reading research papers, so that you might get results best for you.
Have a Clear Goal
Depending on your goal, you should pay attention to different things from the text.
Say you are an 2nd undergraduate student:
- you are not *experienced in the field
- you are reading the paper to get a “gist” of the discussion for your lab
- you don’t care about diving into the methods and the math behind the paper
Say instead you are a 4th year student writing a thesis:
- you are reading a paper to find an answer for your thesis
- you are diving into the deep model structure
- you are scanning thoroughly, but not the exact parameters of the paper
And finally, say you are a phd student building your own simulation following the paper:
- instead, you are searching for those exact parameters, model description
- you do not care AT ALL about the significance of this paper, and other sections like it
Understand that this is the same paper, but because each person has different goals, they all have completely different experiences/outcomes with said paper.
Inner Filter
Regardless of how you find your paper(recommended, discovered), you want to make sure you actually want to read it.
Scientific papers nowadays appear at such a fast rate that it would be impossible to choose without being at least a little picky about what you invest your time into, as you will be putting a lot of time into it.
Papers are tricky:
- titles are mislead
- abstracts are misleading
- you might realize this isn’t something you are interest in midway through
This is why you need to have your mental filter on at all times:
if you find that you no longer are interested in what you are reading halfway through, just ditch it!
Don’t be afraid to go and read something else more interesting instead - your time is valuable! And if you are not interested in what you are reading, you wont absorb the information like you are hoping to.
Given this, in what order should you read this paper?(its parts, etc))
Structure of a Paper
Every paper follows roughly the same structure
Abstract
Abstracts are meant to condense roughly the entire paper into a paragraph, giving you a top-down summary of the findings and process behind the paper
Ideally, abstracts are meant for you to get the “gist” of the paper. However, oftentimes, they only really seem to make sense once you have familiarized yourself with the contents of the paper.
Don’t be afraid of not understand the paper after you have read the abstract a few times, and it still looks like alien garbage. It doesn’t mean you are stupid, just return to it after you have gone through some of the other parts a bit later. Eventually, it will become clear.
Introduction
Contains background information about a papers’ content, necessary prerequisite knowledge that the authors think should be restated in the readers mind.
Sections
The main part of the paper, it can sometimes be just one main part, but is often times quite a few of them.
Each section contains a description of the experiment, detailed explanations of the methods used, and the results of the experiment. Many sections also contain figures and tables.
Discussion/Summary/Conclusion
Names may vary for this section, but this serves as the “wrapping it all together” section.
Authors use this section to summarize what they have done, what it all means, and the results, and how to interpret the findings/significance.
References
Pretty self-explanatory, but don’t underestimate this section.
They will become essential when you are trying to find something briefly mentioned in the text, a good place to go looking for more information.
Order of Reading
Once you have set up your Inner Filter and a Clear Goal in mind, you are ready to get your hands dirty.
- Read the Abstract, skim the Discussion and Introduction
- see if this is something you would genuinely like to read.
- Read the Introduction and Discussion, should be a “deep skim”
- google and research things you don’t understand in the paper, your goal is to get a full picture from this brief.
- watch some youtube videos/explainers if topic is complex
- keep your mental filter on while doing this, try to see the big picture of authors approach