How does dblp handle arXiv publications?

> Home > F.A.Q.

dblp processes arXiv metadata in the following way:

dblp only indexes arXiv papers which belong to a cs.* subject category (as "The Computing Research Repository", CoRR).
Once per month, all relevant publications of the past month are extracted from arXiv's web pages and are included to the bibliography in bulk and in their current state at that point of time. We label that table of contents as a pseudo-issue like "CoRR, May 2013".
Any revisions made to a publication in arXiv after that point will not carry over to dblp. This means, unfortunately, that dblp currently cannot handle "live" documents properly. Keeping our metadata in sync with any potential changes at arXiv is something that we do not have the resources to commit to.

It is important to understand that from the perspective of bibliographic metadata, papers published on arXiv are an independent publication, even if they are closely linked to other publications hosted at, say, a commercial publisher (e.g. a "full version" or even an identical copy). For example, it is quite common for dblp to list the same research contribution in both the arXiv version and the version of the conference proceedings and/or journal article. Such articles can be cited individually and require their own metadata; they are never merged in dblp.

By default, all arXiv publications are understood as preprints and are therefore labeled as "informal publications". If a paper turns out to be part of a conference proceedings or journal series that just happens to be hosted at arXiv (e.g. as an "overlay journal"), then the type labeling will be adjusted as soon as we become aware and the corresponding series' volume is processed.

manage site settings

To protect your privacy, all features that rely on external API calls from your browser are turned off by default. You need to opt-in for them to become active. All settings here will be stored as cookies with your web browser. For more information see our F.A.Q.