Examples
Worked examples
- Is an instance
A network producing template manuscripts in medical imaging where dozens of papers share the same underlying skeleton, with author names, gene targets, and figure colours rotated between submissions.
- Is an instance
A cluster of manuscripts flagged by an editor because they arrived from unrelated institutions but share near-identical phrasing, structure, and a citation pattern to unrelated journals — triggering a formal integrity review rather than an immediate rejection or accusation.
Counter-examples
Looks similar, but isn't
- Not an instance
A professional English-language editing service that improves the writing of an author-composed manuscript without altering its scientific content is not a paper mill.
- Not an instance
A single retracted paper, or a paper that fails a plagiarism check, is not on its own evidence of paper-mill origin — paper-mill identification rests on converging pattern evidence (shared templates, coordinated submission timing, recycled figures across unrelated papers) assessed by an editor or integrity team, not a single red flag.
Editorial commentary
Paper mills are commercial operations that produce fabricated or templated manuscripts and sell authorship positions on them, or that broker the acceptance of such manuscripts into journals — sometimes by compromising guest editors or exploiting inadequately vetted special issues. They have shifted over the past decade from small-scale, individual operations into industrial-scale enterprises capable of producing thousands of manuscripts a year. Coverage of specific paper-mill networks or investigations on this page is described in general, evidential terms — a paper-mill finding is the outcome of a formal editorial or institutional investigation, not an accusation made against a named party on the strength of pattern-matching alone.
Why detection is hard
No single signal is conclusive on its own. Editors and integrity teams instead look for converging evidence: sudden submission spikes from unrelated authors sharing a common broker or agent; recycled figures, tables, or datasets reused across manuscripts that claim unrelated experiments; authors unable to answer basic questions about their own stated methods when asked directly; and manuscripts that share an underlying template — the same structural “skeleton” with author names, gene or disease targets, and superficial details substituted between papers. Because paper mills actively adapt to detection tooling, the specific signals in active use shift over time, and no fixed checklist should be treated as complete or as sufficient grounds for a public accusation on its own.
Tortured phrases and image duplication
Two of the most consistently useful automated signals are tortured phrases — nonsensical synonym substitutions produced by machine paraphrasing used to evade plagiarism detection (e.g. “counterfeit consciousness” standing in for “artificial intelligence”) — and image duplication, where figures or panels are reused, spliced, or duplicated across unrelated papers. Both are covered in full, including their detection tooling, on their own dictionary entries; this page does not restate that detail. Screening tools built to surface these patterns — including the Problematic Paper Screener, a public tool maintained by Guillaume Cabanac and colleagues — post their findings to PubPeer as a triage signal for editors and integrity officers to investigate, not as a self-executing verdict.
The STM Integrity Hub
The STM Integrity Hub is shared, cloud-based screening infrastructure run by STM (the International Association of Scientific, Technical and Medical Publishers) that multiple publishers use as pooled infrastructure rather than each building bespoke detection tooling from scratch. It incorporates donated detection tools from individual publishers — Springer Nature’s Geppetto (an internal tool estimating the likelihood a manuscript section contains AI-generated or tortured-phrase text) was donated to the Hub in April 2025. COPE and STM jointly coordinate United2Act, a multi-stakeholder initiative bringing publishers, integrity officers, and independent sleuths together across five working areas: education and awareness, post-publication corrections, research on paper mills themselves, trust-marker development, and shared terminology.
Institutional response to an allegation
When a paper-mill allegation reaches a journal or institution, the standard response follows a formal, evidence-based process rather than an immediate public accusation: the concern is logged and triaged (often against COPE’s flowcharts for suspected misconduct or duplicate/redundant publication); the named authors are given an opportunity to respond and to answer method-specific questions; where warranted, the institution’s own research-integrity process is engaged in parallel with the journal’s investigation; and the eventual outcome — correction, expression of concern, or retraction — is decided only once that process concludes. COPE’s 2025 update to its Retraction Guidelines specifically addressed batch, paper-mill-driven retractions, including guidance on crediting the third-party sleuths who often first raise the concern and on giving potentially uninvolved co-authors a chance for individual re-review rather than retracting an entire cluster indiscriminately.
References
- COPE & STM, “Paper Mills: Research Report” (2022)
- United2Act consensus statement (2023) and ongoing coalition, united2act.org
- STM Integrity Hub, stm-assoc.org
- Problematic Paper Screener, dbrech.irit.fr (Cabanac, Labbe & Magazinov)
- COPE Retraction Guidelines v2 (2019) and the 2025 paper-mill-focused update
Also known as
manuscript mill · publication mill
Machine-readable encodings
Use in your systems
<role vocab="credit"
vocab-identifier="https://casrai.org/dictionary/"
vocab-term="Paper mill"
vocab-term-identifier="https://casrai.org/dictionary/term/paper-mill" />{
"@context": "https://schema.org",
"@type": "DefinedTerm",
"@id": "https://casrai.org/dictionary/term/paper-mill",
"name": "Paper mill",
"identifier": "https://casrai.org/dictionary/term/paper-mill",
"description": "A commercial operation that produces fabricated or low-effort manuscripts and sells authorship slots, or that brokers acceptance of such manuscripts into journals, typically at industrial scale. A submission is described as originating from a paper mill when the underlying research was not conducted by the named authors, was fabricated to order, or was assembled from a shared, reused template — an assessment made through pattern evidence and formal investigation, not a label to attach to an individual paper on suspicion alone.",
"inDefinedTermSet": "https://casrai.org/dictionary/domain/research-integrity#set",
"url": "https://casrai.org/dictionary/term/paper-mill",
"sameAs": [
"manuscript mill",
"publication mill"
],
"license": "https://creativecommons.org/licenses/by/4.0/",
"publisher": {
"@id": "https://casrai.org/#organization"
},
"author": {
"@id": "https://casrai.org/#editorial-team"
},
"datePublished": "2026-05-21T02:22:52",
"dateModified": "2026-08-22T12:16:34",
"inLanguage": "en-GB",
"isAccessibleForFree": true
}
