Is the number of accepted papers normally determined before papers are submitted? Or is it determined based on the quality of the papers?

I'm primarily interested in CS conferences.

2 Answers 2


In my experience the Program Committee often has a target rate or a target number. Conferences occur on a fixed number of days, and given the length of the talks you'd like to have and the number of parallel sessions or rooms available, you can back your way into the number of papers you can accept.

  • Is this information sometimes publically available?
    – pir
    Commented Nov 20, 2015 at 22:29
  • 2
    Sometimes. You can also just ask the PC or look at the size of past years' versions of the conference.
    – Bill Barth
    Commented Nov 20, 2015 at 22:31
  • @BillBarth, or look at the tentative schedule for a rough idea.
    – vonbrand
    Commented Nov 25, 2015 at 11:58
  • 1
    target rate or a target number — But even those target numbers are really target ranges. For many years, my main conference had a target range of 40-45 accepted papers, but the actual number of accepted papers varied between 35 and 55, depending on the quality (and to a smaller extent, the quantity) of the submissions.
    – JeffE
    Commented Nov 25, 2015 at 18:41

My experience (computer architecture conferences) is almost universally the latter: The percentage of papers accepted is a function of the quality of the submissions (relative to the quality bar for that conference), not the number of submissions. This usually yields the expected result, although over time it may force changes in the conference schedule (e.g., if the research community is growing). Occasionally, the General Chair may have to adjust the conference schedule if the result is not what was expected (e.g., half- vs. full-day schedule on the last day, fewer/more parallel sessions, etc.)

You must log in to answer this question.

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