Recent statistics indicate that the number of retracted scientific articles has risen to unprecedented levels, reinforcing concerns about a broader “reproducibility crisis” across many disciplines. Two contrasting phenomena underlie this trend: paper mills and post-publication peer review.

Paper mills are organizations that produce fabricated manuscripts for profit. They have contributed substantially to the growth in retractions. These entities systematically generate fraudulent studies, often bypassing conventional peer review safeguards, and thereby undermine confidence in the scholarly record.

Unlike traditional peer review, which occurs before a paper is published, post-publication peer review (PPPR) invites the wider scientific community to reassess published work, identify errors, and flag potential misconduct after the fact.

A community of dedicated volunteers, frequently referred to as “sleuths,” has formed around this activity. Much of their work takes place on platforms such as PubPeer, where users may post comments, either anonymously or under their own names, on any article. PubPeer currently hosts more than 300,000 comments, making it a central venue for ongoing scrutiny of the literature.

In September 2024, a meeting of PPPR practitioners was held in Paris under the title “Decontamination of the Scientific Literature.” Participants discussed recent developments and came to agree that:

1. Undetected problematic papers contaminate large swaths of the scientific literature

2. Systematic scientific fraud appears to be increasing

3. The rate of detection is limited by the number of active reviewers

4. Expanding participation in post-publication peer review is essential

5. Non-specialists can contribute meaningfully

6. Barriers to entry persist

Upon this consensus, we initiated the Collection of Open Science Integrity Guides (COSIG). COSIG is an openly licensed, continuously expanding repository of practical guides authored by publication-integrity specialists and experienced PPPR practitioners. At the time of writing, COSIG comprises 31 guides that cover a range of topics, organized into five categories: basics of PPPR; general guides; biology and medicine; materials sciences and engineering; mathematics, statistics and computer science. The guides are intended to lower the threshold for participation, providing clear, step-by-step instructions that enable individuals with diverse backgrounds to engage in PPPR activities. By offering a shared body of knowledge, COSIG aims to alleviate the number of people participating in PPPR and to foster a more inclusive, sustainable community of literature stewards.

Visit cosig.net

Watch a video introducing COSIG and some of its contents, presented by COSIG maintainer Reese Richardson at the King’s College London Open Research Summer School in 2025:

Guides Currently in COSIG

Implications for the broader scientific enterprise

The issues addressed here have relevance beyond the immediate community of PPPR volunteers. A robust post-publication review system contributes to restoring public trust in science, especially when high-profile retractions attract media attention. Funding agencies and journal editors are increasingly acknowledging the value of PPPR as a complementary quality-control mechanism, which may influence future editorial policies and grant-evaluation criteria. Moreover, the open-access nature of modern scholarly publishing means that non-academic readers now have the same opportunity to examine primary research as professional scientists. COSIG equips these readers with the methodological tools necessary to assess the credibility of scientific claims, thereby promoting a more informed public discourse.

Other uses of COSIG?

Beyond its use for PPPR, we hope that COSIG will serve as useful material for rigor and reproducibility courses and as a valuable reference for institutional integrity officers investigating integrity concerns.

Future directions

The Paris consensus emphasized two strategic priorities: (1) scaling up the number of individuals involved in PPPR, and (2) reducing the practical obstacles that discourage participation. COSIG represents a concrete step toward achieving both goals. As additional guides are developed and the community expands, it is anticipated that the detection of problematic papers will accelerate, leading to a gradual reduction in the prevalence of fraudulent publications.

In summary, the combination of an active PPPR community, the widespread use of platforms such as PubPeer, and the availability of open-source instructional resources like COSIG together constitute a promising framework for enhancing the integrity of the scientific literature.