How to get your paper cited?

But, there are literally thousands of papers published each year. How do you make sure that people actually read and cite your work? 🥺
Sharing some tips I learned from others below.
❌ Dump all the details of your work.
✅ Tell a story of a research problem, what people in the past tackled it, why it remains unsatisfactory, and how your work fits in this story to fill a specific knowledge gap.
Write a good related work ():
✔️ Organize the literature to help people learn quickly.
✔️ Contextualize your contributions.
No laundry list please!

I enjoy reading/writing the related work section of a paper. It helps organize prior research and put the contributions of the work in proper context.
But HOW? Check the thread below👇
No one likes to implement your work from scratch before they can improve on top of it.
✔️ Share code, data, (intermediate/final) results.
✔️ Provide examples on how to extend your work.
✔️ Build a demo to quickly try things out.
Found a new paper and disappointed that they did not cite the relevant works of yours? 😥
They may not be intentional. Send a note to the authors and politely ask if they could include the citation. Most would be happy to do so.
❌ Publish the code and leave it as is for ages.
✅ Maintain open-source research code so that it can reach many people. See the lessons learned from @junyanz89 youtu.be/wRMKaBEReCQ?t=…
Don’t stop at publishing and presenting your paper at conference.
Actively promote your work. No one will do that for you. For example, check out this excellent thread on how to disseminate your work by @jbhuang0604!
Remember that you so much time
• just to build and run the baseline?
• train baseline models on new datasets?
• collect examples that expose the limitations of previous work?
Share all these to help people follow up with your work.
(This one is so good I have to add it to the main thread!)

