Preface

Paper is usually the first publication of a new technology, algorithm, programming method or software tool. By reading papers, we can learn about the latest technological advances and keep our skills and knowledge up to date.
At the same time, essays provide the opportunity to develop an in-depth understanding of a specific topic. They often contain detailed theoretical analysis and experimental results, which can provide a deep understanding of a concept or technology. However, how to read papers effectively is a critical but rarely talked about skill. At first, most people learn by themselves through trial and error, but are often discouraged by frustration. This article introduces a method of reading papers- the three-pass reading method .

three times reading method

The key idea is that you should read your paper a maximum of three times, rather than reading from the beginning to the end. Each pass accomplishes a specific goal and builds on the previous pass:

first time

Browse quickly to get a general idea of ​​the paper. You can also decide if you need to do more reading. This pass usually takes about 5 to 10 minutes and includes the following steps:

1. Read the title, abstract and introduction carefully
2. Read chapter and section titles but ignore everything else
3. Read the conclusion
4. Browse the references and make a mental note of what you have read

By the end of the first pass, you should be able to answer these five Cs (Category, Context, Correctness, Contributions, Clarity):

1. Category: What type of paper is this? Is it a measurement paper? Is it an analysis of an existing system? Is it a description of a research prototype?
2. Context: What other papers does it relate to? What theoretical foundations are used to analyze the problem?
3. Correctness: Does the hypothesis appear valid?
4. Contribution: What is the main contribution of the paper?
5. Clarity: Is the paper well written?
   With this information, you may decide not to read any further. This may be because the paper does not interest you, or you do not know the field well enough to understand the paper, or the author made invalid assumptions. The first pass is sufficient for papers that are not in your area of ​​interest, but may be relevant in the future. If a reader cannot understand the key points of your paper after five minutes, the paper will most likely never be read.

Second time

On the second pass, read the paper more carefully but ignore details such as the proof. It can be helpful to jot down key points or make notes in the margins as you read.
Look carefully at charts, graphs, and other illustrations in your paper. Pay special attention to graphics. Are the axes labeled correctly? Do the results show error bars so that the conclusions are statistically significant? These common mistakes will make the difference between a rushed, shoddy job and a truly excellent job.

Remember to mark relevant unread references for further reading (this is a great way to understand the context of your paper). The second pass should take up to an hour. After this pass, you should be able to grasp the content of the paper. You should be able to summarize the main arguments of your paper to others and provide supporting evidence.

This level of detail is appropriate for a paper that interests you but is outside your research area. Sometimes you don’t understand the paper even by the end of the second pass. This may be because the topic is new to you and has unfamiliar terms and abbreviations. Or the author used a proof or experimental technique that you didn’t understand, so that much of the paper was incomprehensible to you.

The paper may be poorly written, with unsubstantiated assertions and many forward citations. Or maybe it’s just because it’s late at night and you’re tired. You can choose now to:
(a) put the thesis aside and hope that you don’t need to understand the material to be successful in your career
(b) come back to the thesis later, perhaps after reading the background material
(c) Stick with it and read it a third time.

The third time

To fully understand a paper. The key to the third pass is to try to virtually reimplement the paper: that is, recreate the work by making the same assumptions as the author. By comparing this recreation to the actual paper, you can easily identify the novelty of the paper, as well as its hidden flaws and assumptions.

This time requires a lot of concentration. You should identify and challenge every assumption in every statement. Additionally, you should think about how you would express a particular idea yourself. Actual vs. virtual comparisons can provide great insight into proof and presentation techniques in papers, and you’ll likely be able to add these techniques to your toolbox. During this pass, you should also jot down ideas for future work.

This pass will probably take about four to five hours for a beginner and about an hour for an experienced reader. By the end of this pass, you should be able to reconstruct the structure of the entire paper from memory and be able to identify its strengths and weaknesses. In particular, you should be able to point out implicit assumptions, missing citations to related work, and possible problems with experimental or analytical techniques.

Finally, a literature review is conducted

When conducting a literature review, your paper reading skills will be tested. This will require you to read dozens of papers, possibly in an unfamiliar field. Which papers should you read? Here’s how using the three-pass reading method can help: First, use an academic search engine such as Google Scholar, along with some carefully chosen keywords, to find three to five recent papers in the field.

Read each paper once to understand the work, and then read their related work sections. You’ll find thumbnail summaries of recent work, and maybe you’ll be lucky enough to find a recent review paper. If you can find a review like this, you’re done.

Read the review and consider yourself lucky. Otherwise, in the second step, shared citations and duplicate author names are found in the references. These are the key papers and researchers in the field. Download key papers and put them aside. Then, visit the key researchers’ websites to see where they have published recently. This will help you identify the top conferences in the field, as the best researchers usually publish at the top conferences.

The third step is to visit the websites of these top conferences and browse their recent proceedings. A quick glance can often identify recent, high-quality, relevant work. These papers, along with the papers you previously set aside, form the first version of your investigation. Read these papers twice. If they all cite a key paper that you didn’t find before, get it and read it, iterating as needed.