TL;DR
Get business pricing on tech for your team
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
arXiv will cap each person’s submissions at two papers per calendar month from Oct. 1, 2026, while keeping its existing limit of three active submissions. The platform says rising volume, including a sharp increase in AI papers, is straining volunteer moderation; the cap is described as a temporary measure.
arXiv will limit each submitting author to two papers per calendar month starting Oct. 1, 2026, introducing a new restriction as the preprint server faces a steep increase in submissions and mounting demands on its volunteer moderators. The rule applies across categories, but co-authors who did not submit a paper are not subject to the cap, according to arXiv.
The new monthly limit is separate from arXiv’s existing rule, introduced in 2024, that permits an author to have no more than three active submissions at a time. Rejected submissions still count toward the two-paper monthly allowance, because they also require moderation work. Authors who delete a paper before publication can reclaim that submission slot, according to the report.
Submission totals cited in the report show the scale of the increase: arXiv received 9,869 papers in September 2016, 20,569 in September 2024 and 40,363 in September 2026. The September 2026 figure was the highest for any month in the platform’s history. The report says the rise generated nearly 9,000 support tickets for staff and moderators, while submissions in the cs.AI category grew more than sixfold in two years.
arXiv links the pressure in part to AI tools and describes the cap as a temporary measure while it develops moderation capacity. The report says moderators have encountered thin or narrowly focused papers, “salami” papers split into small pieces, and dense text written with AI assistance. These descriptions are concerns reported by moderators, not a claim that all recent submissions share those characteristics.
How the Cap Changes Submitting
The policy places a direct limit on how frequently an individual can send work to arXiv, a widely used preprint server where researchers share manuscripts before or alongside formal peer review. For authors who submit several papers each month, the cap could require prioritizing work or coordinating who submits a co-authored paper. The rule is attached to the submitting author, not every listed author, so another co-author may submit a paper without being counted against a colleague’s personal allowance, according to the report.
The platform’s reason for acting is the labor involved in screening and processing a growing number of papers. Thomas G. Dietterich, chair of arXiv’s Editorial Advisory Council, said that a relatively small share of authors submit many low-quality papers and use a disproportionate amount of moderator time. He argued this can delay other work, including papers from authors submitting quality research. The cap is intended to limit repeated submissions that add to that workload, though its effect on overall volume has not yet been established.
The decision also reflects a broader operational challenge for research publishing: AI tools can help people produce and format text, but volume alone does not establish whether a paper is sound or useful. arXiv’s rule is an administrative response to moderation demand, not a new judgment about the validity of AI-assisted research or a replacement for peer review.
AI research paper submission software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Earlier arXiv Rule Changes
The monthly cap follows other steps to manage computer science submissions. In November 2025, arXiv tightened requirements for review articles and position papers in that area, requiring them to go through peer review, according to The Decoder. In May 2026, Dietterich announced a one-year ban for authors whose papers clearly contained unverified outputs from large language models, such as hallucinated citations.
The workload issue is also appearing at research conferences. The report says ICLR 2027 had received about 50,000 abstracts before its deadline, compared with roughly 19,500 valid submissions to ICLR 2026. It also cites a GPTZero review that found at least 100 fabricated citations among about 5,000 papers accepted at NeurIPS 2025, despite each paper having at least three reviewers. These examples indicate pressures across research review, but they do not show that the same policies or problems apply uniformly to arXiv.
arXiv is also changing its institutional structure. After 25 years connected to Cornell University, it began moving toward independence as a nonprofit on July 1, 2026. In late September, the Simons Foundation International, XTX Markets and the Siegel Family Endowment pledged a combined $17.2 million over three to five years. The report says some funding is intended for technical upgrades, including work related to AI-generated content.
“A relatively small proportion of authors are submitting a large number of low-quality papers and consuming a disproportionate fraction of the moderators’ time.”
— Thomas G. Dietterich, chair of arXiv’s Editorial Advisory Council
As an affiliate, we earn on qualifying purchases.
How Long the Limit Will Last
arXiv has not set a date for ending or reviewing the cap in the information provided by the report. Dietterich has said he hopes it can be increased or removed once arXiv’s moderation tools can scale, but the report does not specify what technical changes are planned, how progress will be measured, or when the organization would decide whether to revise the rule.
It is also not yet clear how the cap will affect submission totals, moderation delays or authors’ choices about which co-author submits a paper. The cited counts establish a sharp rise in submissions, but do not isolate how much of that increase is caused by AI tools. Nor does the report provide a detailed breakdown of rejected submissions, the number of authors who would exceed the new monthly limit, or how exceptions and technical edge cases will be handled.
research moderation management software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Implementation Begins in October
The two-paper limit is scheduled to take effect on Oct. 1, 2026. Authors will need to account for the rule across arXiv categories and remember that rejected papers count, while a paper deleted before publication restores its slot, according to the report. The platform’s existing three-active-submission cap will remain in place.
After implementation, the key indicators will include submission and support-ticket volumes, moderation delays and whether arXiv changes the rule as its technical systems develop. The organization has described the cap as temporary, but no formal timetable for a policy review or a decision to lift it has been reported.
As an affiliate, we earn on qualifying purchases.
Key Questions
When does arXiv’s two-submission limit start?
The new cap is scheduled to begin on Oct. 1, 2026.
Does the limit apply to every co-author?
No. According to the report, the rule applies to the person who submits the paper, not to co-authors who are listed on it but do not submit it.
Do rejected papers count toward the monthly cap?
Yes. Rejected submissions count because they still require moderation work. Authors who delete a paper before publication can reclaim that slot, according to the report.
Does the new rule replace arXiv’s active-submission limit?
No. The existing limit of three active submissions at a time, in place since 2024, remains alongside the new monthly cap.
Is the two-paper cap permanent?
arXiv has described the limit as a temporary fix while it works to scale its moderation tools. Dietterich hopes it can later be raised or removed, but no review date or end date has been reported.
Source: rss
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
