This post contains some common Q&A for PhD students:
Q: Where can I look for papers to read and study?
A: In the beginning of your PhD, start by looking at the list of accepted papers at top- and mid-level conferences. You can do this by googling the conference name, a year and search for a link of ‘accepted papers’ or ‘program’. [Names of conferences: CRYPTO, EUROCRYPT, ASIACRYPT, PKC, Theory of Cryptography Conference] Some conferences also provide video recordings of the presentations (which is a time-efficient way to get what the paper is about). Following this strategy you will most likely encounter solid results and work of good quality.
Once you become more familiar with your field, you can start to look at not peer-reviewed sources, such as:
- eprint (this is the main go-to source for crypto and security papers),
- arxiv (this contains a variety of papers that fall under the computer science umbrella)
- HAL (this is a French-based database, it collects works on various topics within computer science).
¡Important! Not all papers you find on these websites are of good quality. It is important that you read the articles in a critical way. Often you can find full versions of papers accepted at top conference, these contain further details than the ‘official’ published version (conference proceedings).
Towards the end of your PhD (e.g., the last 6-10 months), you should have identified one or two topics you really like to work on. A good habit to acquire would be to periodically check what is new on eprint (e.g., have a look at the feed of the past week / month, so that you are always up-to-date with what is going on in your research field).
Q: How do I know if a conference is good or not?
Time and experience. The main factor is how good the reviewers are, usually top rank conference get better reviewers that will most easily spot weak results. However this is not always the case (sometimes good papers appear also at lower ranked conferences because of the fierce competition). You can check out this list of conference deadline and my ranking for them (A > B+ > B- > C). The same conference might be ranked differently according to the field (crypto vs security vs privacy vs applications). A general reference point is the Australian CORE ranking portal.
Q: Am I allowed to submit the same paper to multiple venues?
A: In general NO, not simultaneously. The large majority of the conferences and journals in our filed only accept original works (i.e., papers that do not have large overlap with existing work). If your paper is rejected from venue1, of course you can submit (an improved version) to venue2, after you receive the final notification of rejection. Always read the call for papers (CFP) in detail, especially if you are a novice. There are a few exceptions: some journals, local conferences without proceedings and workshops accept works that have already been published or ‘full version’ (usually at least 25% new material) of existing publications.
Q: Am I allowed to reuse (verbatim) text?
A: NO, this is plagiarism even if the text in question comes from your previous paper. There are two exceptions to this rule:
- You are working on an extended version of your own work, e.g., when you want to publish to a journal you often include more content than the proceeding version of the article that appears at a conference. This is a very special case, and often requires you to write a statement of what is new with respect to the already published version (and the novelty threshold is around 25-30%).
- Definitions, Syntax, Security Models (usually in Sections Background, and Preliminaries). Here it is good practice to reuse well-established pieces of text. Of course, you need to cite the (already published) paper you used as a source. Sometimes people skip the citation if the definition is soooo standard, e.g., Discrete Logarithm Problem/Assumption, definition of Bilinear Pairings etc. See this paper for an example.