Paper Guidelines

From Mw_writing
Jump to: navigation, search

As with all types of writing, standard guidelines apply. Additionally, the following guidelines apply.

General Paper Guidelines[edit]

Papers are meant to communicate ideas, but their format brings with it certain conventions.

Know your contribution[edit]

This is applicable to all types of writing, but is especially applicable when you are competing to get your writing accepted. The purpose of a paper is to share the work you have done with the academic community in your field. You should be able to clearly explain (in person) what your contribution is. If you are unsure of what your contribution is, your paper will most likely show a lack of focus, which will make it unlikely to be accepted.

In a related vein, you should know what problem you are solving. If it is not immediately obvious to a reader what problem you are solving, you probably should explicitly state it.[1]

Follow formatting requirements[edit]

Most publications have requirements for how submitted papers must be formatted. Typically, a LaTeX template will be provided. It is a good idea to use this template, as it will allow you to easily follow the formatting guidelines. Some people [2] say that breaking formatting requirements just makes it more likely that your paper will be rejected, while others [3] say that this is almost a sure way to get rejected. If you want to get published, it's probably not worth the risk.

Use a conventional structure[edit]

Typical academic papers have a number of sections, with each section having a specific set of information in it. One possible structure[4] is

  • Title: Clear, possibly catchy, name that describes your work.
  • Authors, Date, Keywords: Location and style of these are typically dictated by the conference or journal you are submitting to.
  • Abstract: Addressed below, an abstract serves to give a very brief summary of your work in 100 to 250 words. You abstract will be reproduced independently of your paper, so it should stand on its own.
  • Acknowledgments: Thank those who have helped you, including grants and other funding sources. This can also go at the end of your paper.
  • Introduction: Introduce the problem to your audience, and explain why it is important.
  • Background or Related Work: Summarize what has been done previously related to this problem, and explain why the previous papers do not solve this problem as well as you do.
  • Resources and Methods: Explain how you studied or solved the problem.
  • Results: Explain and discuss what you found out, and what you are contributing to the field.
  • Discussion: Discuss the meaning of your results.
  • Conclusion: Re-state what your contribution is, and possibly discuss potential areas of future work.

This is only one possible structure. The content in this outline is present in almost every paper, but the exact order and naming of the sections varies from paper to paper. If in doubt, take a look at papers previously published in the journal or conference you are interested in.

Look at other papers[edit]

Look at other papers from the journal or conference you are interested in submitting to. See what the other papers say, how they say it, what sources they cite, and in what order they say what they say. Determine which of these characteristics are important to mimic.

Don't rush the Related Work section[edit]

The related work section is a very important section of your paper. This section establishes the context of your work, proves to readers that your work is a genuine contribution, and convinces reviewers that you understand the field you are trying to contribute to. You should see if there are any seminal works relevant to your work, and see if citing them makes sense.

You should also check out the past few years of publications at the venue you are submitting to. If you've chosen your venue wisely, work relevant to you should have been published there previously. You should assume that the reviewers are very aware of these papers, and make sure you refer to relevant ones.

Edit mercilessly[edit]

If your paper is accepted for publication, it will be forever linked with your name. Recruit some friends and colleagues to read your paper and find all the problems (or areas needing improvement) they can. If you don't ask them to find errors, the reviewers will.


Abstracts typically have a very tight word limit, often around 200 words. In the abstract, you are supposed to convince the papers committee that you have a useful contribution. Abstracts are also commonly reproduced in libraries and paper indices. Many researchers will first look at the abstract to determine if the paper itself is relevant to their interests and worth reading.

Completely summarize your contribution OR Hint at your full contribution[edit]

This is an area of debate, with some academics preferring to leave a nugget of their contribution out of their abstract, while others feel that the abstract should be sufficient for a reader to know what it is that your paper is contributing. If you address all your contributions in your abstract, it makes it much easier for someone reviewing the literature to quickly decide whether it is worth their time to read your paper.[5]

Avoid references in your abstract[edit]

Your abstract will be reproduced separately from your full paper, and as such should be completely self-contained. If you must make reference to previous work, mention it by author instead of by footnote number.[5]


  1. Roy Levin and David D. Redell: An Evaluation of the Ninth SOSP Submissions, or, How (and How Not) to Write a Good Systems Paper
  2. Mary-Claire van Leunen and Richard Lipton: How to Have Your Abstract Rejected
  3. Jim Kajiya: How to Get Your SIGGRAPH Paper Rejected
  4. Dan Hyde: How To Write a Research Paper
  5. 5.0 5.1 Henning Schulzrinne: Writing Systems and Networking Articles