Conference Paper Or Talk: Fill & Download for Free

[A2A]I assume you mean “What is the difference between a journal paper and a conference paper?”The answer varies from field to field. In most fields, the answer is simple: Journal papers count, and conference papers don’t. Computer science is a bit of an outlier, in that conference papers actually count as real publications. So here are some differences between the two types of papers in CSLJournal papers are typically longer than conference papers. (For largely historical reasons, most conferences have page limits. Most journals do not.)Journal papers are typically more polished than conference papers.Journal papers typically take significantly longer to publish than conference papers. Conferences run on a fairly tight submission-reviewing-decision schedule, with the submission deadline four to six months ahead of the actual conference, and with the complete conference proceedings available either shortly before or shortly after the conference. Journals have no such schedule; as a result, it is not uncommon for journal submissions to take multiple years to be reviewed, revise, re-reviewed, copy-edited, and finally published.Most journal papers are more stringently reviewed than most conference papers, but the difference is less than you might think. (I’ve received several conference-paper reviews that were longer and more detailed than a typical journal referee report.)Fewer people read journal papers (because they take too long to publish, and the field moves fast).Fewer papers are submitted to journals (because the audience is smaller).Publishing in a typical journal is easier than publishing in a typical conference. Conferences are limited by the number of talks that can be given during the 3–5 days of the actual meeting; journals are only limited by the size of their hard drives.All else being equal, journal papers have less influence on your professional reputation than conference papers (because fewer people read them).As a result, journal papers have less influence on hiring, tenure, and promotions decisions than conference papers, at least in strong departments (because professional reputation matters far more than the number of bullets in your CV).

I am assuming this question is asking about the comparison of conferences vs arXiv, in terms of visibility. Because conference papers are peer-reviewed, an average paper at a good conference would be better in quality than an average paper on arXiv. As such, seasoned researchers pay a lot more attention to conference papers — they spend several days attending a conference, listening to most of the talks, and going to the posters of most papers that interest them. An average arXiv paper is unlikely to get that much attention from a top researcher. Often, the researchers only look at arXiv papers that are very similar to the problem they are working on, or papers from other well-known researchers in the field. So a not-so-famous researcher will probably not have a good reach to top researchers via arXiv. The advantage of arXiv, on the other hand, is that even if your paper does not qualify as a conference paper, you can still upload it on arXiv, and get a reasonable visibility, particularly among junior researchers, including graduate students. Also, uploading on arXiv puts out your work in the community much earlier than through conferences.A lot of conferences these days allow you to upload your work on arXiv, while it is being reviewed. In that case, you get the best of both worlds — faster release to community, visibility to junior researchers, visibility to senior researchers [if accepted at the conference].Coming to the other aspect of the comparison — building your profile — again, because conference papers are peer-reviewed, it is harder to publish a conference paper than an arXiv paper, and therefore, having conference papers adds to the credibility of your work.

In my field of CS/AI, papers in top peer-reviewed conferences are every bit as much respected as papers in peer-reviewed journals. The conference papers are how we communicate in real time, with sites such as arXiv or informal circulation of pre-prints increasingly enabling communication in “real real time”.Journals, increasingly, are where research ideas go to be entombed or enshrined, years after everyone has read about the idea and reacted to it. Or the journals serve as an outlet for ideas that don’t fit into the short conference-paper format — maybe a paper combining ideas from a number of conference papers.Of course, journals (often) have a stronger system of peer review, and informally circulated work has not been formally peer reviewed at all. That function is gradually being taken over by informal communication networks: good papers get buzz within the community, and people hear about them.This transition is weighed down by the academic promotion process, which in many places still depends on counting people’s peer-reviewed journal papers. Older, more traditional fields are more tied to the idea of journals, and when a tenure case comes up at the university level, there is often an argument with these people about whether a reputation build upon more modern forms of communication is really legitimate. The better the university, the more they try to count ideas rather than journal papers.So, yes, in CS we definitely do list conference papers on our CVs, right along with any journal papers. And we even list posters at good conferences, if the full paper was published in some archival form and if the paper attracted some attention (or should have).If we have to cut down the publication list to fit some arbitrary limit, the deciding factor is usually which papers were most influential (citation count correlates with this, though not perfectly), and not whether is was published in a journal or conference.