Before you start reading, consider why you want to read this paper and what I want from it. You need to control the way you read. If you just need to summarize, just skim it. If you want to introduce this paper to others, you need to read it deeply and doubt the argument until you fully understand it. If you will use this information in the future, taking notes will help you remember it. If you don't know what you want from it, you don't know whether reading is beneficial or a waste of time.

Preparatory work/about to start work

A quiet place, pencil, paper or a copy of paper.

Read what?

Read headlines and abstracts

Read the full text, archive or skip.

read extensively

What did they do?

Skip introduction, title, chart, definition and conclusion.

Consider credibility and usefulness.

(who wrote it, whether it is famous, where it works, and what prejudice does the unit have;

Where the paper is published, the popularity of the magazine, whether it is a censored academic journal;

When was it written? Is it out of date? )

Browse references

(Is it widely quoted? Does the author know the latest works? Have you consulted the classic papers on this subject? Have you read some papers in some references? Do you know they didn't cite relevant research? )

Decide whether to continue reading further.

Deep reading

When you read this paper in detail, treat it with scientific skepticism. You can do this by trying to break the argument.

How did the author do it?

Doubt the author's argument

Test hypotheses (whether their results depend on hypotheses and whether these hypotheses are reasonable)

Test methods (can they measure what they claim? Can they explain what they have observed? Do they have enough control? Is the experiment conducted in a standard way? )

Test statistics (is a reasonable statistical test applied? Have you made a reasonable error analysis? Are the statistical results meaningful? )

Test reasoning and conclusion (can the conclusion be drawn from the observation results in a logical way? Is there any other explanation for the observed phenomenon? Other conclusions and correlations that they didn't point out in the data? )

How can I apply their methods to my work?

take notes

Taking notes will help you understand what you read and save time in the future. When you read a paper, you understand it well. Clear definition, icon association at a glance. You will be clear when you write a report on this topic next week, or when you need to refer to this paper next year.

Read and write.

Highlight key parts

(Key points or standard underline on the edge; When you find a new term, abbreviation or definition, write def next to it. When you find an example to illustrate a point, take notes on the side;

When you see an icon, test it, what does it mean and what is its trend? What about correlation? Take a note and explain it in your own way)

Record new terms and definitions

Inductive tables and maps

When you see it related to other papers, take notes beside it. If you doubt an argument, write down your objection. If you know a suitable quotation, write it down)

Set an example for yourself.

This will tell you whether you understand the definitions and terms, and make you deeply understand why you get a rule/principle or result. What are the examples in the paper?

Write a summary

When you read a paper in depth, write a short summary. State in your own words what you have learned from this paper. What's your theme? Put the abstract and the paper together for reference.