Skip to content
OctymPowered By Proofig AI
For publishers

Every integrity and quality check, in one review, configured for each journal you run.

From submission screening to post-publication investigation, for your editorial and integrity desks alike. One platform, configured to each journal's own requirements by your Octym team.

What every review covers

One review, every lens.

Whatever a journal chooses to lead with, the same three families of check are available on every submission. Your configuration decides what runs and what leads.

Internal manuscript integrity

  • AI-generated text signals
  • Image manipulation
  • Image duplication
  • AI-generated images
  • Graph & chart duplication
  • Statistical & analytical consistency
  • Figure–legend consistency
  • Claim & citation consistency
  • Internal consistency
  • Prior publication & reuse disclosure

External checks & requirements

  • Image & graph plagiarism
  • Retracted reference detection
  • Hallucinated / non-existent references
  • Incomplete & broken references
  • Reference analytics
  • Data & code availability
  • Paper-mill signals
  • Authorship & affiliation consistency
  • Funder requirements
  • Submission requirements & completeness

Research quality

  • Methods & rigor
  • Reproducibility
  • Replicability signals
  • Research ethics
  • Novelty & contribution
  • Peer-review-grade analysis
Two desks, one platform

Editorial and integrity teams each get their own Octym.

Different desks, different jobs, different review profiles, all on one platform, without separate logins.

Editorial offices

Editors · managing editors · editorial assistants

Every submission screened against your journal's own rules before review time is spent, including the paper-mill signals worth catching ahead of peer review: the major risks first, each linked to its evidence, exportable for the record.

Integrity teams

Integrity managers · post-publication reviewers

When a reader raises a concern about a published paper, start from a full evidence-backed review, with every finding traced to its source, shaped for a defensible decision.

Triage
major issues first

Major issues

  • Figure 2traced · Fig. 2b

    Possible duplication with Figure 4a.

    High signal
  • Referencestraced · Ref. 14

    Reference 14 was retracted in 2022.

    Elevated

Then review

  • Statisticstraced · Table 3

    Reported p-value differs from the recomputed test.

    Medium
  • Ethicstraced · Methods ¶6

    Ethics statement present and consistent.

    Clear
Each finding traced to its source, exportable for the editorial record.
Per-journal configuration

Every journal is different. So is every review.

No two journals share the same instructions for authors, the same article types, or the same worries. Your Octym team configures each journal's review to its own rules: which checks run, which one leads, and what it is measured against. One title leads with novelty and contribution; another leaves it out entirely and spends that attention on rigor and reproducibility.

One publisher · two journals · two reviews

Each journal gets its own set of checks and its own emphasis.

Journal A

Lives on novelty

  • Novelty & contributionDeep
  • Claim & citation consistencyDeep
  • Reference verificationStandard
  • Image & figure integrityStandard
  • Reporting completenessLight
  • Replicability signalsOff
Journal B

Built on experimental rigor

  • Methods & rigorDeep
  • Statistical & analytical consistencyDeep
  • ReproducibilityStandard
  • Image & figure integrityStandard
  • Reporting completenessDeep
  • Novelty & contributionOff

Same desk. Two journals. Two different reviews.

Adding a journal is a configuration, not a development cycle.

Across the whole life of a paper

The right check, at the right moment.

The same platform follows a manuscript from the day it arrives to long after it's published.

1

At submission

Screened against your journal's rules (structure, figures, references, statistics) before review time is spent.

2

At revision

Run the review again on the revised version, so the figures the authors replaced are checked, not only the ones that were screened.

3

After publication

When a concern is raised, a maximum-depth review of the published paper, with every finding traced to its source.

Complete control

Your configuration is yours.

What runs, what leads, and when it changes. You decide; your Octym team makes it happen.

You decide

Which checks run for each journal, what leads the review, how deep it goes.

Your Octym team builds it

A new journal, a new rule, a new focus. Tell them, and it's configured for you.

It changes when you do

Instructions for authors evolve, article types appear, worries shift. Your review keeps up.

Image & figure integrity is powered by Proofig AI, a trusted leader in scientific image integrity, adopted by leading journals and publishers.

Journal review profiles are configured with you at onboarding, each to the journal's own instructions for authors. Adding a journal later is a configuration, not a project.

FAQ

Questions, answered.

Octym screens every submission across every integrity and quality lens and returns one evidence-linked review your editorial and integrity teams can act on, designed to complement your existing process, not replace it.

Yes. Each journal gets its own review profile (its own checks, its own emphasis), configured by your Octym team to that journal's instructions for authors. Two journals at one publisher can run completely different reviews.

You do. What runs, what leads, and when it changes stays under your control. You decide, and your Octym team makes it happen.

No. It surfaces signals with evidence, with every finding traced to its source. Decisions stay with your editors and integrity officers.

Image and figure integrity is powered by Proofig AI, a trusted leader in scientific image integrity, adopted by leading journals and trusted by the world's top researchers, publishers, and research institutes.

Octym screens every submission for image problems, statistical inconsistencies, paper-mill signals and missing statements, against your journal's own rules. Octym can review the entire paper across three categories of checks: internal manuscript integrity, external checks and requirements, and research quality, and each journal's profile sets what runs and what leads. Figures get forensics powered by Proofig AI, references get checks powered by RefGuard, statistics are tested against each other, and claims are read with their citations. Editors see the major risks at the top, each linked to its evidence.

Yes, Octym surfaces the signals that cluster in paper-mill submissions, before peer review time is spent on them. Each comes from its own check: figures reused from other papers, powered by Proofig AI, the largest image-plagiarism database in the industry; authorship and affiliation inconsistencies; undisclosed reuse of related work; and missing disclosure of the original data files behind key images. Point tools each see one slice. Octym reads these signals together, each with its evidence, never as a conclusion about the authors.

Octym is built so a journal can screen every submission while the authors' unpublished work stays confidential. Manuscripts and grant applications are analyzed on secure, private servers, and Octym never uses them to train AI models. That holds for Octym's own models and its AI providers', which are contracted on terms that prohibit training on what they receive. Processing is automated, staff cannot browse or open submissions in the ordinary course, data is encrypted in transit and at rest, and sharing is off by default.

Not for integrity screening: a general-purpose AI chatbot reads text patterns and guesses, while Octym reads the whole submission against the published record. A chatbot is fast and helps with language, but it cannot run image forensics, can fabricate citations, leaves no evidence trail and is a real confidentiality risk for an author's manuscript. Octym adds validated image checks powered by Proofig AI, statistics tested against each other and every finding traced to its source, with an audit trail. Editors decide.

No, Octym is designed to complement peer review and your editorial process, not replace them: it surfaces signals, and your editors make every call. Before reviewers are invited, it runs the checks your journal's profile sets and adds a reviewer-style Assessment, major issues leading, each comment anchored to the passage it is about. Octym is not an automated accept or reject machine. Every finding links to its evidence and exact location, and when an editor confirms, edits or resolves a finding, that decision is logged to an audit trail.

Because a revision can bring replaced figures, and running Octym again checks the revised version, not only the one screened at submission. Octym covers three stages on one platform: at submission, before review time is spent; at revision, so the replaced figures are checked; and after publication, with a maximum-depth review when a concern is raised. Each journal's profile sets the depth for each stage, such as a fast screen at submission and a forensic pass at revision.

References that do not support the claims they are attached to are one of the signals that cluster in paper-mill submissions. Octym flags them with reference checks powered by RefGuard, Proofig AI's reference integrity product, along with citations to retracted papers and references that cannot be matched to a real publication. Editors see them beside Octym's other paper-mill signals, such as figures reused from other papers, before peer review starts. Each is a signal with its evidence, never a conclusion about the authors.

See it on your journals.

Tell us how your journals review. We'll configure Octym around them.

Analyzed on private servers. Never used to train models.