Receiving reviewer comments can feel like a confirmation that your manuscript is close to publication—until you see requests for new citations. A reviewer suggests adding three references, a coauthor forwards two PDFs they found in their reference manager, and you realize you must confirm every citation exists, is correctly formatted, and supports the claim you are attaching it to. This guide walks through a complete verification sequence for citations added during peer review, with checkpoints you can reuse for any revision.
The core challenge is not finding more papers. It is proving that each added citation maps to a real, retrievable source and to the exact sentence it is meant to support. This article covers how to establish that trail before you submit a revised manuscript. Along the way, it shows where WisPaper fits: finding candidate papers, screening BibTeX entries, and organizing verified sources so your revision stays auditable.
What Is Your Starting Point?
Before verifying anything, identify what kind of addition you are handling. Most revision citations fall into one of three groups.
Reviewer-requested citations. A reviewer names specific works or asks you to cite a research area. You need to locate the actual papers and confirm they are relevant to the point the reviewer raised.
Self-added citations. You or a coauthor insert references while addressing a comment. These often arrive as pasted BibTeX, a DOI, or a PDF from a reference manager. The risk is that metadata was copied incorrectly.
Edited citations. A coauthor changes the volume, page range, or author order of an existing reference. Even if the paper is real, the citation may no longer match the published record.
Your first checkpoint is to list every new or changed citation in a revision log. For each entry, record the in-text location, the claim it supports, and what metadata you currently have. Do not skip this step. When a journal asks for a point-by-point response, this log becomes the evidence that your citations are deliberate rather than decorative.
Where Should You Look First?
Start with the journal's own expectations. COPE frames peer review as a process that depends on trust, and that trust extends to the references you add after review. The reviewer is not required to verify your citations for you, so your revised submission should make verification straightforward.
The ICMJE recommendations describe the standard parts of a biomedical manuscript and emphasize following the selected journal's current instructions. Confirm whether your target journal requires a specific reference style, a separate reference list, or particular identifiers for cited works. A citation that is real but formatted incorrectly can still create a submission delay.
Once you understand the format, search for the exact paper behind each new citation. Use the metadata you have as a starting point: a DOI is ideal, while a partial title plus author may require several searches. When you need to locate papers through a scholarly search interface, you can use WisPaper search to find candidate records and compare them against what your coauthor pasted into the manuscript. Keep the search results visible while you check each field.
How Do You Confirm a Citation Is Real?
The simplest verification test is to find the paper in a database record and match its metadata. Check these fields in order:
- Title. Watch for missing words, changed subtitles, or translated titles.
- Authors. Confirm first authors, initials, and the number of coauthors.
- Journal and year. Verify the volume, issue, and page range against the published version.
- DOI. If the citation includes one, confirm it resolves to the same paper, not a similar title.
A mismatch in any field is a signal to investigate, not an automatic failure. Sometimes the database record is correct and the citation in your manuscript was mistyped. Other times the paper exists but the metadata came from an unverified export.
For BibTeX entries added by collaborators, consider a targeted screening step. TrueCite checks BibTeX against real academic databases to flag hallucinated references. This is useful when a coauthor sends a .bib file with several additions and you want a systematic screen before manually reviewing each entry. After running the check, open the flagged items and compare them against the database records you found in your search.
How Do You Stop Hallucinated References From Entering the Revision?
Hallucinated references are a particular risk when a coauthor uses a conversational tool to draft response letters or suggest readings. The citation can look complete—author, year, journal, DOI—while pointing to a paper that does not exist or to the wrong study.
Protect the manuscript with a two-layer rule. First, require every new citation to include one verifiable identifier or a source you can retrieve directly. Second, never accept a citation until you have opened the paper or its database record yourself.
A common failure pattern looks like this: a coauthor adds a reference with a plausible title and a DOI that fails to resolve. If the DOI does not resolve, search for the title alone. If the title search returns no matching paper, remove the citation from the manuscript and note the removal in your response letter. If the title search returns a slightly different paper, update the citation to the real record and tell the coauthor what changed.
This is also where your revision log earns its place. When you remove or correct a citation, write the original text, the corrected text, and the reason. Reviewers often ask how you handled their suggested readings, and a clean log shows you did more than copy a reference list.
What Does a Verification Workflow Look Like End to End?
A complete workflow includes discovery, screening, checking, and storage. Here is a sequence you can adapt to your manuscript.
1. Capture the request
Put every reviewer-requested or coauthor-added citation in one spreadsheet or project note. Include the comment that prompted it, the proposed citation, and the sentence where it will appear.
2. Search and retrieve
Find the paper in a scholarly database. For each proposal, keep the search query and the retrieved record together. If you are using WisPaper, Research Projects gives you a place to group the revision work and keep related searches and verified sources in one project rather than scattered across email threads.
3. Screen metadata
Compare the proposed citation against the retrieved record field by field. For BibTeX additions, use TrueCite as a first-pass screen before manual checking.
4. Verify relevance
Open the paper and confirm it actually supports the sentence you are attaching it to. A real citation in the wrong place is still an error. Save or upload the verified papers to My Library so the final set of revision sources stays together and accessible to coauthors.
5. Update the manuscript and log
Make the edits, then record the final citation, its source, and the resolution in your revision log.
When Should You Use a Shared Working Project?
Multi-author revisions create their own verification problem: different coauthors add citations in different working copies, and no single person sees the full set. A shared project reduces duplicate searches and keeps the final reference list traceable.
Create one shared project for the revision. Inside it, store the reviewer comments, the papers each coauthor proposes, and the final verified list. When a coauthor suggests a new source, they add the paper to the shared project instead of editing the manuscript silently. This makes it obvious which citations are new, who added them, and whether they have been screened.
If you are coordinating through WisPaper, creating a project under Research Projects gives you a dedicated area for this revision. You can collect candidate papers, run the screening step, and keep the verified sources in the project so the next coauthor can see the same material.
How Do You Check That the Final List Is Submission-Ready?
Submission readiness is not the absence of errors you happened to notice. It is a completed checklist across every new or modified citation. Work through this table before you send the revised manuscript.
| Checkpoint | What to do | Pass condition |
|---|---|---|
| Source exists | Retrieve the paper or its database record | You have the paper open or a matching record saved |
| Metadata matches | Compare title, authors, journal, year, volume, pages, DOI | Every field matches the published record |
| Identifier resolves | Click the DOI or ISBN | It opens the intended paper or book |
| Claim alignment | Read the sentence and the cited passage | The source directly supports the stated claim |
| Format compliance | Check against the journal's instructions | Reference style and required identifiers match |
| Revision log updated | Record the original proposal and final citation | The change is traceable by any coauthor |
| No unverified additions | Compare the reference list against the log | Every new or changed citation has a log entry |
Run the checklist once for reviewer-requested citations, once for self-added ones, and once for edited existing references. For the edited group, pay extra attention to the metadata match step, since a changed page range or author order can be hard to notice in a long reference list.
How Do You Document the Method Trail for Reproducibility?
Across disciplines, research quality increasingly means leaving a transparent record of how findings were produced. For citation work, that record is your revision log plus the search queries and retrieved records behind each addition. NIH frames scientific rigor as strict application of the scientific method, including transparent reporting of design, methods, analysis, and interpretation. A citation trail is a small but concrete part of that transparency.
A minimal method trail includes:
- The reviewer comment or coauthor request that triggered each citation
- The search query or database used to locate the paper
- The retrieved record compared against the proposed citation
- The final accepted metadata and its source
- The date and person who verified each entry
This record matters even if the journal does not request it. It lets you answer a later query in seconds, gives coauthors a shared reference point, and protects the manuscript from silent edits during future rounds. Store the trail where all coauthors can reach it—your shared project or an internal revision file—rather than in a personal inbox.
What Does a Worked Example Look Like?
Suppose you are revising a methods-focused manuscript, and Reviewer 2 writes: "The authors should cite recent work on reproducibility in computational pipelines."
Step 1: Capture. You log the comment and create three rows in your revision sheet, one for each candidate reference you will consider.
Step 2: Find and screen. You run a scholarly search for recent reproducibility work. You find two strong papers and one plausible title forwarded by a coauthor. When you screen the coauthor's BibTeX through TrueCite, that third entry is flagged. You search for its title and find the actual paper has a different volume and year than the BibTeX claimed.
Step 3: Verify. You open the two strong papers and confirm they directly discuss protocol transparency. For the corrected third paper, you check the published version and decide it is relevant but slightly dated. You keep all three, with corrected metadata for the third.
Step 4: Link and store. You add each verified paper to your My Library collection for this revision and attach the in-text locations where each citation will appear.
Step 5: Edit and respond. You update the manuscript, then write in the response letter: "We added three citations on reproducibility in computational pipelines. We verified each against the published record; one coauthor-suggested reference had incorrect metadata and was corrected before inclusion."
If a coauthor later asks why the third citation differs from what they sent, your revision log has both versions side by side.
What Does It Mean to Finish?
You are finished when all three of these conditions are true.
- Existence: Every new and modified citation in the manuscript has been retrieved and matched against a database record or the published paper itself.
- Alignment: Every added citation sits next to a sentence it directly supports, and you can point to the supporting passage if a coauthor asks.
- Traceability: Your revision log maps each change from the reviewer request or coauthor proposal to the final reference, with enough detail that another person could repeat the verification.
Do not treat the response letter as the finish line. The letter summarizes the work; the verification trail is the work. If your final manuscript reference list matches your verified project records one-for-one, the citation portion of the revision is ready. For papers you want to interrogate further before finalizing the revision, you can ask questions based on the papers saved in your library through Library QA.




