论文阅读技巧

Paper types

• Propose / formulate a new problem;
• Solve a problem people care about;
• Exposition and survey.

In computer science

• Theory / algorithm / system / application
• Tools. service and infrastructure.

You must make your paper easy to read.

You’ve got to make it easy for anyone to tell

• what your paper is about
• what problem it solves
• why the problem is interesting
• what is really new(and what isn’t)
• why it’s so neat

And you must do it up front.

Tile and levels of summarization

• Title(one sentence)
  • TL; DR(few sentences)
• Abstract(4-8 sentences)
  • Extended abstract(one page)
• Introduction(1.5 pages)
• Paper(8 pages, double-column, including 1 page of reference)

Do Geman: A paper has four parts. Spend equal amount of time writing each.

1. Title
2. Abstract
3. Introduction
4. The rest of the paper

Reader cares more about the first two pages(1, 2, 3)(maybe conclusion)

• Is this paper worth reading?

• Should l know something about this paper?

• Can this be quickly rejected(by conference)?

• 图文阅读：弄清楚论文中的表格和图片的标题和注释，就能获得这篇论文八九成的信息。

• 了解论文的类型： 1、提出问题型论文；2、解决问题型论文； 3、阐述和调查型论文。

Critical reading: ask questions, be suspicious

If the authors attempt to solve a problem

• are they solving the right problem?
• are there simple solutions not have considered?
• what are the limitation of the solution?

If the authors present data

• do they have right data to substantiate argument?
• do they gather it in the correct manner?
• do they interpret data in a reasonable manner?
• are there other more compelling data sets?

Are the authors assumptions reasonable?

• Is the logic clear & justifiable, given assumption?
• Is there a flaw in the reasoning?

Creative reading: positive thinking

• What are the good ideas in this paper?
• Improvements that might make important practical differences?
• Applications/extensions that the authors might not have thought of?
• Can they be generalized further?

If you were going to start doing research from this paper, what next thing would you do?

Write a one-page review

Your summary:

• Key idea in a sentence or two.

Learning from critical reading:

• any problems?

** Learning from creative reading:**

• idea for next year’s conference?

You opinion:

• do you recommend this paper?

Share this review with others?
Do a presentation?

Analyze with a list of questions

• What’s main claim? Key idea?
• What’s key limitation?
• Is there code available? Data?
• Is the idea neat? Is it counter-intuitive?
• Is the contribution significant enough?
• Is the experimentation good? Manual tuning?

---

• Are there important related papers l missed?
• Is it useful to my work e.g. product dev?
• Is this a paper worth following?
• What do other people think about this paper?
• Who are dominating this area of research?
• What question should l ask the authors?

Tools to help reading/ what to do when stuck?

• Google/ Bing Web search
• Google scholar/ Semantic scholar/ Microsoft academic search
• ArXiv and other sites for papers
• OneNote? Taking notes!
• Cliffsnotes, sparknotes for research?
• Mendeley? Reference management.

Tips

**Write a summary, do a presentation about paper.
Notes:

• bib, summary, key points, figures, citations, structures, styles

Ask:

• people around you, more experienced researchers, the authors.

Read more, Keep reading.

Be patient:

Analyze with a list of questions

• What’s main claim? Key idea?
• What’s key limitation?
• Is there code available? Data?
• Is the idea neat? Is it counter-intuitive?
• Is the contribution significant enough?
• Is the experimentation good? Manual tuning?

---

• Are there important related papers l missed?
• Is it useful to my work e.g. product dev?
• Is this a paper worth following?
• What do other people think about this paper?
• Who are dominating this area of research?
• What question should l ask the authors?