arXiv Changes Rules After Getting Spammed With AI-Generated 'Research' Papers (404media.co) 11
An anonymous reader shares a report: arXiv, a preprint publication for academic research that has become particularly important for AI research, has announced it will no longer accept computer science articles and papers that haven't been vetted by an academic journal or a conference. Why? A tide of AI slop has flooded the computer science category with low-effort papers that are "little more than annotated bibliographies, with no substantial discussion of open research issues," according to a press release about the change.
arXiv has become a critical place for preprint and open access scientific research to be published. Many major scientific discoveries are published on arXiv before they finish the peer review process and are published in other, peer-reviewed journals. For that reason, it's become an important place for new breaking discoveries and has become particularly important for research in fast-moving fields such as AI and machine learning (though there are also sometimes preprint, non-peer-reviewed papers there that get hyped but ultimately don't pass peer review muster). The site is a repository of knowledge where academics upload PDFs of their latest research for public consumption. It publishes papers on physics, mathematics, biology, economics, statistics, and computer science and the research is vetted by moderators who are subject matter experts.
arXiv has become a critical place for preprint and open access scientific research to be published. Many major scientific discoveries are published on arXiv before they finish the peer review process and are published in other, peer-reviewed journals. For that reason, it's become an important place for new breaking discoveries and has become particularly important for research in fast-moving fields such as AI and machine learning (though there are also sometimes preprint, non-peer-reviewed papers there that get hyped but ultimately don't pass peer review muster). The site is a repository of knowledge where academics upload PDFs of their latest research for public consumption. It publishes papers on physics, mathematics, biology, economics, statistics, and computer science and the research is vetted by moderators who are subject matter experts.
Not quite (Score:5, Informative)
That's not actually what the announcement [arxiv.org] says:
(my emphasis). They're still accepting preprints of research papers without prior peer review.
Having a dearth of sympathy here (Score:2)
I guess it's difficult to be particularly sympathetic when a publication or service that is rather important-to and dependent-upon the research that has brought us what they're currently calling AI is swamped with low-quality, low-effort garbage generated by the same sorts of systems that their service has enabled.
I'm still wondering how/why this happens if one presumes that the names of researchers are associated with any academic papers. It seems like the reputational harm that should be done to someone
Re: (Score:2)
There are enough John Doe without any university affiliation that have not much to lose when arXiv (or even publishers) ban them. If everyone with "I have an opinion and ChatGPT wrote an article about it" only uploads three papers before being banned, that are still a lot of papers.
Re: (Score:2)
Is that what's happening? My impression was these papers come from academics hoping to pad their CVs. If so, even a single instance should be made a career-killer. There needs to be a registry of people who've done this.
Re: (Score:1)
Re: (Score:2)
There are already plenty of junk journals that will publish anything they're paid to for anyone who wants to bump their publication count. None of them are
Stop posting 404media slop (Score:3, Informative)
No, arXiv does **not** change the rules for research papers.
They change the rules for review articles and position papers.
Here is the original link: https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Fblog.arxiv.org%2F2025%2F10... [arxiv.org]
And the main quote:
> Is this a policy change?
>
> Technically, no! If you take a look at arXiv’s policies for specific content types you’ll notice that review articles and position papers are not (and have never been) listed as part of the accepted content types. Review articles and position papers have, in the past, only been accepted at moderator discretion, because the few we received were of high quality and of interest to arXiv readers and the scientific community at large.
You can still post your research. But if you post an article that is only a review of other papers (what is a valid publication, but no own research) arXiv wants to know that its something someone would publish and not just someone asking ChatGPT or Claude for a summary of papers relevant to a topic.
And for position papers ... I think you don't need LLMs to produce shitty position papers. ArXiv is not your soap box, so make sure to have something backing your position or publish it on your own blog.
Bring back the WoT! (Score:3)
Spam, spam, spam, eggs and spam didn't provide enough incentive to try to distinguish between humans and skin jobs, but now "AI slop" does? Ok, great!
Check the OpenPGP signature.
Unsigned? /dev/null.
Signed but no trust path? /dev/null
Signed and with a trust path? Can still be trash, but its claims to be of human origin, are worth taking seriously. If you find a problem (e.g. someone trusted the wrong person) then deal with that then.
This article was AI generated slop (Score:3)
The irony is, this article itself was AI-generated slop with ridiculous duplication. Maybe a low-effort AI-assisted piece by an author who couldn't be bothered.