iThenticate flagged a new submission at 18% overall, with dense clusters in Methods and a few phrases in the Discussion, all from the authors’ prior preprint. I can justify overlap in protocols, but the paraphrasing looks mechanical and the attribution is patchy — two figures cite the preprint, one does not. Where do you draw the line at initial check, and do you request revision before peer review or proceed with a note to reviewers?
the line at initial check, and do you request revision before peer review or proceed with a note to Agree on the timing part — what’s worked for me is to proceed but place a 48-hour hold asking for a one-paragraph text-recycling statement, fix all figure captions to cite the preprint, and either cite or fully rephrase the Discussion snippets; if I see any >3 consecutive sentences from the preprint without citation, I kick it back pre-review.
My take: I’d lean toward the simplest next step and see if it changes anything this week — if not, you’ve got a clear case to escalate. What would block you from trying that?
I’ll proceed at screening if it’s <20% and mostly Methods, but I send a 48‑hour “reuse map” request — one-page table listing what’s from the preprint, where it’s rephrased, and fixing all figure attributions; if the Discussion copy reads mechanical or they can’t supply the map, I kick it back before review. @jameson_lee92’s timer works, and the map keeps us from playing whack‑a‑mole.
Quick example: I ask for a one‑line provenance note at the start of Methods and a caption audit for any reused figures (with DOI), then proceed; if the preprint’s license is ND or the Discussion reads like templated paraphrase, I pause for a brief rewrite before review. @micclar, does that match what you’re seeing?
I run a doc‑to‑doc iThenticate against their own preprint to see if the clusters are truly just protocol text; if it reads like a thesaurus took the wheel, I ask them to swap paraphrase for a plain line like “We used the previously reported protocol (DOI: …)” and to add the DOI in the caption and on the figure file itself. I’ll proceed on that basis, but if any Discussion language reuses interpretations (not just boilerplate), I pause and request edits first.