November 9, 2026

How to Answer a Reviewer Who Requests More References

A reviewer's request for more references is rarely a demand to pad the bibliography. It is usually a signal that the reviewer could not see the scholarly grounding for a claim, a method, or a comparison. Your job is to m

Written byWisPaper TeamAI Research Workflow Team
Editorial illustration for How to Answer a Reviewer Who Requests More References

A reviewer's request for more references is rarely a demand to pad the bibliography. It is usually a signal that the reviewer could not see the scholarly grounding for a claim, a method, or a comparison. Your job is to make that grounding visible, verifiable, and proportionate — then to show the reviewer, in the response letter and the revised manuscript, exactly what changed and why.

The practical path has five moves: decode what the reviewer is actually asking, decide which claims need support, find and verify candidate works, integrate them where they do work, and leave an auditable trail your coauthors can follow. This guide walks through each move with checkpoints, a worked example, and completion criteria you can apply before you resubmit.

What is the reviewer actually asking for?

Before you search for anything, classify the request. Most reference requests fall into a few recognizable types, and each calls for a different response.

  • Missing foundational support. A claim is stated without any citation, or with a citation that does not carry the weight placed on it.
  • Missing recent work. The reviewer knows of newer studies that extend, qualify, or contradict what you wrote.
  • Missing competing or contrasting views. Your literature framing omits a line of work the reviewer considers relevant.
  • Missing methodological precedent. A procedure, measure, or analysis choice needs a source the reader can follow.
  • Citation quality concerns. A cited work may be misattributed, outdated, or not actually support the sentence it is attached to.

Read the reviewer's comment literally. If they name specific works, those are your first candidates. If they describe a gap without naming works, your task is to identify the claim in your manuscript that created the gap and find support that genuinely addresses it. A response that adds references unrelated to the flagged sentence will read as deflection.

Checkpoint: Write one sentence per reviewer comment stating the claim under question and the type of support requested. Keep this list; it becomes the backbone of your response letter.

Which claims actually need a new citation?

Not every flagged sentence needs a new reference. Some need a rewritten sentence. Decide this before you search, because it controls how much work you are committing to.

For each flagged claim, ask:

  1. Is the claim load-bearing? Does the argument depend on it, or is it background? Load-bearing claims need the strongest support.
  2. Is the existing citation adequate? If a cited work does support the sentence, the fix may be to make the connection explicit in the prose rather than add another reference.
  3. Is the claim too broad? A sweeping statement may need to be narrowed to what the cited evidence actually shows, which can resolve the request without new sources.
  4. Does the reviewer want breadth or depth? A request for "more references" on a contested topic usually means representing the range of positions, not stacking citations on one side.

Record your decision for each claim: add support, narrow the claim, replace the citation, or explain the existing citation more clearly. This decision log is the artifact that keeps the revision honest and makes coauthor review fast.

How do you find candidate references efficiently?

Once you know which claims need support, search deliberately rather than broadly. Two complementary approaches work well together: targeted searching for known items, and exploratory searching for work you have not yet identified.

For known items — works the reviewer named, or papers you half-remember — search by author and title, and use date limits and sorting to narrow results. Google Scholar's documented features include searching by author and title, date sorting and limits, Cited by and Related articles links, citation export, and ways to locate alternative versions of a work. Those Cited by and Related articles links are especially useful when you have one strong anchor paper and want to find the work that builds on it.

For exploratory searching, use a scholarly search interface to run topic queries and screen results by relevance. WisPaper search lets you find academic papers through a scholarly search interface, which is a reasonable starting point when you need to map a topic before committing to specific sources. If you prefer a public entry point that explains the search capability first, the WisPaper Scholar Search page serves that purpose.

When a reviewer's request is broad enough that you need a structured literature pass rather than a handful of lookups, a longer research task may fit better. The Scholar Agent area is designed for longer research tasks with literature review use cases, which suits the situation where a reviewer asks you to engage with a whole body of work.

Checkpoint: For each flagged claim, you should have a shortlist of candidate works with enough metadata to judge relevance. Do not yet commit them to the manuscript.

How do you verify that a reference is real and actually supports the claim?

This is the step where reference requests go wrong. A reference that exists but does not support the sentence is worse than no reference, because it invites a second round of review.

Verification has two independent parts: confirming the work's identity, and confirming that its content supports your claim.

Identity. A DOI resolves through doi.org to the current destination registered for that identifier, which helps you locate a work — but resolution alone does not confirm that the bibliographic fields match what you intend to cite. Check title, authors, venue, and date against the source itself. Crossref exposes publisher-deposited scholarly metadata through its REST API, including titles, authors, venues, dates, identifiers, licenses, funding, ORCID and ROR identifiers, abstracts when deposited, and post-publication updates. That makes it a useful cross-check for bibliographic fields, while the source text still needs to be read for claim support.

Claim support. Read the relevant passage in the candidate work and confirm it says what your sentence says it says. Note the specific section or finding you are relying on. If the support is partial, narrow your sentence to match.

If you maintain a BibTeX file, screening it before submission catches entries that do not correspond to real records. TrueCite checks BibTeX against real academic databases to flag hallucinated references — a useful pass when your bibliography has grown through many hands or many revision rounds.

Checkpoint: Every new reference in your shortlist should have a verified identity and a noted passage supporting a specific sentence. Anything that fails either test comes off the list.

Where should the new references go in the manuscript?

Adding references is an editing task, not an append task. Each new citation should land where it does work.

  • In the introduction or framing, new references should sharpen the gap statement or represent the range of prior positions.
  • In the methods, a citation should let a reader reproduce or evaluate a choice.
  • In the discussion, references should connect your findings to specific prior results, including ones that complicate your interpretation.
  • In a response to a breadth request, consider whether a short paragraph engaging with the literature serves the reviewer better than a string of citations attached to one sentence.

As you place each reference, update the sentence around it. A citation dropped into unchanged prose often reads as an afterthought. Where the reviewer asked for engagement with competing work, write the engagement.

What does a strong response letter look like?

Your response letter is where the reviewer sees your reasoning. Structure each response so it can be checked quickly:

  1. Restate the comment in your own words so the reviewer knows you understood it.
  2. State what you did — added references, narrowed a claim, replaced a citation, or explained existing support.
  3. Point to the location in the revised manuscript.
  4. Name the works added and, in one clause each, why they belong.

If you declined to add a reference the reviewer suggested, say why in terms of claim support or scope, and show what you did instead. Reviewers accept reasoned disagreement far more readily than silence.

Worked example: a reviewer flags an under-supported claim

Suppose a reviewer writes: "The authors claim that this measurement approach is widely used, but cite only one study. The literature is broader and includes contrasting applications."

Step 1 — Classify. This is a breadth request combined with a possible overgeneralization. The claim ("widely used") is broader than a single citation can carry.

Step 2 — Decide. Two actions: narrow the sentence to describe what the cited study actually shows, and add references representing the broader and contrasting applications.

Step 3 — Search. Use the existing citation as an anchor and follow its Cited by and Related articles links to find later work. Run topic searches for the measurement approach to catch applications the anchor paper does not cover.

Step 4 — Verify. For each candidate, confirm identity through the DOI and bibliographic fields, then read the passage that supports the application you are citing. Drop any candidate whose support is tangential.

Step 5 — Integrate. Rewrite the sentence to state the narrower claim, then add a short passage acknowledging the range of applications with the verified references attached.

Step 6 — Respond. In the letter: restate the comment, note that the claim was narrowed, list the added works with a clause on each, and give the section and paragraph where the changes appear.

Step 7 — Record. Log the search queries used, the candidates considered, and the reason each rejected candidate was dropped. This log is what lets a coauthor or a future you reconstruct the revision.

How do you keep the revision auditable for coauthors and future readers?

A reference revision that only exists in your head is fragile. Build a trail as you go.

Keep a working table with one row per flagged claim:

FieldWhat to record
Reviewer commentThe comment, quoted or closely paraphrased
Claim locationSection, paragraph, sentence
DecisionAdd support / narrow / replace / clarify
Search queriesThe queries and databases used
Candidates consideredWorks reviewed, including rejected ones
Reason for rejectionWhy a candidate was not used
Final referencesVerified works added, with identifiers
Response textThe wording used in the letter

This table does double duty: it is your checklist during revision and your evidence if a coauthor or editor asks why a particular work was or was not included.

For organizing the papers themselves, a library workspace helps keep candidates, verified sources, and background reading separate. My Library lets you save or upload papers and organize them, which is useful when a single reviewer comment pulls in a dozen candidates. If you want to ask questions grounded in the papers you have collected, Library QA supports asking questions based on papers uploaded to My Library or added from search results — helpful when you need to check what a collected paper says about a specific application before citing it.

When the revision is part of a larger effort with multiple threads, Research Projects provides a project area where you can create a new project and keep the work grouped.

When is the revision ready to resubmit?

Use explicit completion criteria rather than a feeling that you have done enough. The revision is ready when all of the following hold:

  • Every reviewer comment about references has a corresponding entry in your decision log.
  • Every new reference has a verified identity and a noted passage supporting a specific sentence.
  • Every flagged claim is either supported, narrowed, or explicitly addressed in the response letter.
  • The response letter names each added work and its location in the revised manuscript.
  • The bibliography has been screened for entries that do not correspond to real records.
  • A coauthor who did not do the revision can follow the log and reconstruct each decision.

If any criterion fails, the gap is visible to a reviewer too. Fix it before resubmitting.

A reusable checklist for reference requests

  • Classify each reviewer comment by type of support requested.
  • Identify the specific claim each comment targets.
  • Decide per claim: add, narrow, replace, or clarify.
  • Search for known items by author and title; use date limits and Cited by / Related articles.
  • Run topic searches to map unfamiliar literature.
  • Verify identity: DOI resolution plus bibliographic field check.
  • Verify claim support: read the passage, note the section.
  • Screen the BibTeX file for entries without real records.
  • Place each reference where it does work and update the surrounding prose.
  • Write a response entry that restates, states action, gives location, and names works.
  • Complete the decision log for every comment.
  • Confirm all completion criteria before resubmitting.

FAQs

Go back to the sentence or paragraph the comment is attached to and identify the claim that is under-supported. The request is almost always about a specific claim's grounding, not about the bibliography's overall size. Once you have named the claim, classify the type of support needed — foundational, recent, contrasting, or methodological — and search for that. If the comment is genuinely general, respond by strengthening the framing of your literature review and explain in the letter which claims you reinforced and why.