Academia Stack Exchange is a question and answer site for academics and those enrolled in higher education. It only takes a minute to sign up.
Sign up to join this community
I am a PhD student in computer science and have submitted a paper to a conference which details my proposed architecture. However in the month or so since this paper has been submitted my architecture has undergone changes (relatively major), how is this normally dealt with?
If the paper is accepted do I present the work as it was at the time of submission or do I openly state that changes have been made and present the updated version of the architecture?
I believe you should do what is better to the audience.
In some disciplines/communities it is strongly assumed that conference participants read the abstracts/papers before the presentation, and expect the discussion to follow precisely the material that is published. In this case, of course, you could only comment briefly on the recent changes you've made.
In other fields (such as mine), the abstracts are used only as a means to (roughly) describe the topic of the presentation. It is quite normal to extend the actual talk beyond the scope of the abstract, or reduce the role of some pieces of research advertised in the abstract. In this case, I'd advise you to make a talk that is simply interesting for you and the audience. It can probably include both the original design and your contribution to its recent development. Make sure people in the room understand why the changes were needed, how you have worked them out, and what is the benefits of your new architecture.
I agree with @dmitry-savostyanov in that you should do what's best for your audience.
They've come to hear something interesting and if your work changed, then that seems to indicate that information on why and how it changed would be very relevant to their interests.
In CS, conference papers have a different status from conference papers or abstracts in other fields: they're long, they are actually considered publications, they're usually not published before the event, and most people in the audience will not have read them. So your audience probably won't be surprised to hear something that's not in the paper. On the other hand, if you really pique their interest, they may read your paper and feel disoriented or disappointed.
I would say report on the best results you have while being clear about what they'll find in the paper.
In computer science, you have 2 options:
In case your paper is accepted, you can update the paper for the final camera-ready version. If the changes are not too big, this can be a viable possibility and is actually no too uncommon.
If the changes are too major, you can prepare an extended version of the conference paper for submission to a journal.
