Who is Peer reviewer for?
Students and researchers looking up AI for peer review who must write a referee report on a manuscript they are not allowed to share.
GuidesPath
Peer reviewer
Settle the venue AI rule first, then build the report from one human read and the public citation record — not from a pasted PDF.
IntermediateLessons5
Time~110 min
Tools5 tools
Before you start
- The manuscript and the venue, editor, or course policy
- Enough domain knowledge to judge the method, not only the writing
Tools in this guide
0%0/5 steps
Path artifact
A structured report: summary, major issues with fixes, minor issues, citation flags with public URLs, and a recommendation that follows from the majors.
Finish every lesson and you should have: A structured report: summary, major issues with fixes, minor issues, citation flags with public URLs, and a recommendation that follows from the majors.
Find the venue, editor, or course policy on generative AI in peer review and copy the sentence that covers submitted work. Perplexity may hand you the URL; you open the page yourself. Write one word at the top of your notes: allowed, forbidden, or unclear. Unclear counts as forbidden.
Pitfall
Taking a model's summary instead of the page. Publishers revise these policies mid-cycle, and the version a model learned is last year's.Verification
You can quote the sentence that permits or forbids uploading this manuscript, and name the page it sits on.Read the whole manuscript with the pen down, then write the main claim in two sentences from memory. File it in Zotero with venue, deadline, confidentiality status, and the policy lock from lesson one. The PDF stays on your machine on this pass.
Pitfall
Starting margin comments on page two. You spend the rest of the review arguing with sentences instead of with the design.Verification
You can summarise the main claim in two sentences with the PDF closed.Build a four-column table: claim, evidence offered, method behind it, conclusion drawn. Mark where the chain breaks and write why. Claude may pressure-test your map only if the policy lock reads allowed; otherwise the map stays in your own notes.
Pitfall
Uploading the PDF after you already logged a ban, because the deadline is close and the map is slow work.Verification
You can name the weakest link and the passage in the manuscript that creates it.Take the references the argument leans on. In Scite, see whether later papers cite each one as supporting or contrasting. In Semantic Scholar, confirm the paper exists as described and test the novelty claim against a search. None of this needs the confidential PDF.
Pitfall
Writing “citation needed” from a list you never opened. Some of those references exist and say close to the opposite of what you assumed.Verification
Every citation flag links to a public record a co-reviewer could open.Open with a summary in the authors' own terms, then major issues, minor issues, and a recommendation that follows from the majors. Every major comment names the problem, says why it changes the conclusion, and offers one thing the authors could do. Manuscript text stays out of any cloud polish the venue forbids.
Pitfall
Asking a model to make the tone friendlier until the major issue reads like a suggestion the authors can skip.Verification
Every major comment has an action, and the recommendation follows from the number and severity of the majors.You should now have: A structured report: summary, major issues with fixes, minor issues, citation flags with public URLs, and a recommendation that follows from the majors.
- You can quote the sentence that permits or forbids uploading this manuscript, and name the page it sits on.
- You can summarise the main claim in two sentences with the PDF closed.
- You can name the weakest link and the passage in the manuscript that creates it.
- Every citation flag links to a public record a co-reviewer could open.
- Every major comment has an action, and the recommendation follows from the number and severity of the majors.
Questions this guide answers
Every answer below is assembled from the dated fields on this page. Nothing is written separately for search.
What does Peer reviewer get you to?
Settle the venue AI rule first, then build the report from one human read and the public citation record — not from a pasted PDF.
What do you finish Peer reviewer with?
A structured report: summary, major issues with fixes, minor issues, citation flags with public URLs, and a recommendation that follows from the majors.
What goes wrong in Peer reviewer?
Each lesson names the mistake you are about to make:
- Taking a model's summary instead of the page. Publishers revise these policies mid-cycle, and the version a model learned is last year's.
- Starting margin comments on page two. You spend the rest of the review arguing with sentences instead of with the design.
- Uploading the PDF after you already logged a ban, because the deadline is close and the map is slow work.
- Writing “citation needed” from a list you never opened. Some of those references exist and say close to the opposite of what you assumed.
- Asking a model to make the tone friendlier until the major issue reads like a suggestion the authors can skip.