How to read a paper?

study › skill

怎么读论文

third pass

Goals:

  1. general idea
  2. Content, not details(such as proofs)
  3. Read in depth

first pass

5~10 min, 决定是否继续读

  • 步骤：
  1. 读标题，摘要，介绍
  2. 读部分和小部分的标题，略去其他
  3. 读结论
  4. 瞥一眼引用，去掉读过的
  • 回答问题：5C

  1. Category: What type of paper is this? A measurement paper? An analysis of an existing system? A description of a research prototype?

  2. Context: Which other papers is it related to? Which theoretical bases were used to analyze the

    problem?

  3. Correctness: Do the assumptions appear to be valid?

  4. Contributions: What are the paper’s main contributions?

  5. Clarity: Is the paper well written?

  • 适用类型：可能相关，不在研究氛围内

Second pass

1h

  • 步骤
  1. Look carefully at the figures, diagrams and other illustrations in the paper. Pay special attention to graphs.
    1. Are the axes properly labeled?
    2. Are results shown with error bars, so that conclusions are statistically significant?
  2. Remember to mark relevant unread references for further reading (this is a good way to learn more about the background of the paper).
  • 回答：summarize the main thrust of the paper, with supporting evidence, to someone else
  • 适用类型：感兴趣的，但不在研究范围之内
  • 当仍旧看不懂这篇论文在讲啥：延后再读

third pass

4-5 hours, virtually re-implement the paper

  • making the same assumptions as the authors: identify and challenge every assumption in every statement
  • recreate the work: think about how you yourself would present a particular idea.
  • By comparing this re-creation with the actual paper, you can easily identify not only a paper’s innovations, but also its hidden failings and assumptions.

读什么论文

暂时跟着课表走就好了。都要记下对之后可能有用的点（积累

review

我个人的一个小建议是，看的时候，自己心里也想一想，如果是我来做，我会怎么做？如果我来讲这篇paper，我要怎么样才能给别人讲懂

关于paper review：

  1. review是什么？所有会议paper，在投稿之后，怎么判断中还是不中呢？这个流程叫同行评审，是找几个同行专家对你的paper进行review，了解你paper的内容，判断你paper的优缺点，决定是接收还是录取。

  2. review的主要内容：paper的内容与优缺点。一般是大约半页到一页A4纸的篇幅。

  3. review的目的：这个过程会需要你去 主动地思考paper的内容与价值，批判性地学习与吸收，这样的效果很多时候比 被动地接受 要更好一些。

  4. 如何判断paper的价值？这也是在review的时候你们要问自己的问题。我这里给一些参考建议

    解决的问题是什么？重要嘛？
    效果怎么样？显著嘛？
    表述怎么样，清晰嘛？
    逻辑合理嘛？
    技术是否有问题？
    影响力有多大？
    是否够详细，能根据它复现出来嘛？
    核心的思想是什么，有道理嘛？
    实验是否完整，是否合理公平？

注：方法部分来自How to Read a Paper；review部分来自胡 Sir(胡津铭)的分享和建议。