38
$\begingroup$

I guess most of us know that one can easily automatically generate a math-like nonsense paper, and that it is possible to have such a paper published. However, I was quite sure that nobody actually does that.

But recently I was looking at papers which cite one of my colleague's articles, and I encountered [edited] a paper that I do not want to discuss here, so let me omit the details. Unfortunately I do not really have time to have a closer look at it. (Actually, I tried a bit, but, well, I could not understand a single sentence.) However, the way my colleague's work is cited, complete lack of proofs of theorems, and abundance of buzz words make it sort of difficult to believe that this article was not written in a computer-assisted way.

I do not really want to ask about that particular paper (and of course I might be wrong about it). What I am interested in is a broader picture.


Question. Do automatically generated nonsense articles ever get published in "decent" mathematical journals? If yes, how often does this happen? Does anyone keep track of that?


A quick online search did not yield anything interesting. The Wikipedia entry on "sting articles" only mentions the single paper from 2012 (which was eventually not published because APC have not been paid by the author). This is the same paper which is described under the links that I gave in the first paragraph. There are loads of information about other disciplines (e.g. here), but, as far as I can tell, nothing about mathematics.

$\endgroup$
16
  • 10
    $\begingroup$ It is perhaps worth remarking that the author seems to have been active for many decades, much longer than the AI has been around, and that one of the previous recent papers (ieeexplore.ieee.org/document/5780078) comes with the following note on the publisher's website: "Notice of Retraction: After careful and considered review of the content of this paper by a duly constituted expert committee, this paper has been found to be in violation of IEEE's Publication Principles. We hereby retract the content of this paper. Reasonable effort should be made to remove references to this paper." $\endgroup$ Oct 5, 2022 at 11:04
  • 25
    $\begingroup$ Don't close this post. It is a good idea to improve our awareness about fake journals. $\endgroup$ Oct 5, 2022 at 12:03
  • 7
    $\begingroup$ If you don't want to judge a particular paper, and steer conversation away from it, then why mention the particular paper? It seems to add little to the post except as a focus for a conversation you (I think correctly) don't want to have. $\endgroup$
    – LSpice
    Oct 5, 2022 at 12:39
  • 5
    $\begingroup$ @LSpice: I was hoping this will give a clearer view of what I meant., but I guess you are right. I'll edit the question momentarily. $\endgroup$ Oct 5, 2022 at 12:42
  • 6
    $\begingroup$ Whatever it is, the example you initially provided was written by one of the 3 advisory editors of the journal. So it means it has no peer reviewing (or a phoney one), that the other editors didn't intervene, that the journal does not have the policy (of any decent journal) that editors shouldn't publish (blatant conflict of interest), and that the publisher didn't detect the issue (which is more understandable, since it's the editorial board's responsibility and competence to check that things are meaningful). (...) $\endgroup$
    – YCor
    Oct 5, 2022 at 15:41

1 Answer 1

28
$\begingroup$

The Problematic Paper Screener tries to identify papers and preprints resulting from Algorithmic Text Generation based on engines such as MathGen (for papers in math) and SCIgen (for papers in computer science).

The database lists 24 MathGen entries, some from serious publishers:

The practice seems much more prevalent in computer science, with 265 SCIgen entries, from these publishers:

$\endgroup$
2
  • 2
    $\begingroup$ One of the listed Springer papers is actually a book chapter, and I have acquired a pre-retraction print copy of the book, so it is for real. Unfortunately only a couple of pages of the chapter are Mathgen; most of the rest of it is simply plagiarized. $\endgroup$ Oct 6, 2022 at 4:37
  • 5
    $\begingroup$ For the record: (1) None of the MathGen-generated papers listed at PPS was published in a mathematical journal, which is kind of optimistic. (2) I searched the PPS database for math journals: there is a bunch of them, but not too many, and of course some may be false positives. (3) The paper that I have accidentally found myself is not listed in PPS. The journal where it was published appears once, based on the use of "tortured phrases". (4) It is unclear to me how reliable the filters used by PPS are. There seem to be few false positives (see FAQ there), so there can be many false-negatives. $\endgroup$ Oct 6, 2022 at 8:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.