Assemble your title from the previous conference buzzwords.
Discover a deep result implicit in early work of the technical program chair indicating their prescience.
Become a reviewer and reject all other papers in areas close to your own.
Have an influential list of authors, some of whom are big-shots.
Re-submit a paper without any changes.
Include the potentially most risky reviewers as authors; they can be removed after the paper was accepted.
Cite the most famous work of the program chair even if it is not related to your work.
Kernelise an existing linear algorithm that has not yet been kernelised (becoming difficult as most have been done).
Adhere to the page limits making use of the available tools.
It is well known that the probability of acceptance increases the closer your submission time is to the deadline. What is less well know is that the probability can continue to increase past the deadline. There are two approaches:
Manual Excuses
Submit the paper format instructions instead of your actual paper.
Submit a version of the paper that is clearly incomplete and hope the program chair will kindly ask you whether you submitted the right version.
Advanced Technical means The above excuses are not guaranteed to work, because the program chair may suspect you are deliberately tricking him. Never fear, the JMLG extreme programming team has developed PDPCP (Plausibly Deniable Paper Corruption Program) which will subtlety change the PDF file of your paper to make it unreadable and unprintable, or, for higher credibility, make the particular pages you have not finished unreadable.
Write a clear concise paper that properly addresses an important and relevant problem.