I'm a little confused about the two concepts of contributions and novelty especially when I want to clarify them in my papers (CS scope). For example, as a common practice, people talk about their contributions at the end of the introduction chapter. So, is this the place that I have to emphasize all the novel parts of my algorithm/approach? Or should I talk about how better is the performance of my approach comparing to others?

If I do not have to mention the detail novelty of my method in the contribution section, then where should I clearly emphasize them to make sure the reviewer get them without too much effort?

  • Might depend a bit on what tier conference and how competitive it is. Also, depending on how you achieved the performance improvements, that might be novel in and of itself.
    – anonymous
    Nov 8, 2018 at 16:19

1 Answer 1


It depends. Novelty simply means "something that hasn't been done before", while "contribution" specifically talks about pushing the state of the art and solving a problem that hasn't previously been solved (or solving an old problem better).

To illustrate the point, assume you are working on a big data application and you realize that no one yet has ever tried algorithm X on that problem. You apply algorithm X and try to publish simply because no one has done so before. This is novelty, but it probably wouldn't be considered a contribution unless algorithm X is clearly better and you provide a solid justification for why.

A contribution is all about pushing the boundaries of what is known. To continue our example, let's say now that you produced a new algorithm, Y, and are able to prove that in this specific problem it is better compared to other algorithms. Then this would be both novel and a contribution.

When writing a paper with a "contributions" section, you usually should stick to the contributions in the sense described above - how does this work push the boundaries of the field.

It should be clear from the introduction that:

  • A problem exists in the field that hasn't yet been solved (or that can be solved better)

  • Your work solves/helps that problem (this is the novelty - if your work doesn't do this, then what's the point?)

Then in your contribution section, you can specifically explain what your contribution is ("we propose an algorithm for... that uses... to solve...")

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .