Paste a reference list to identify verified, suspicious, and potentially fabricated citations.
Paste Format Example
1) Warren JR, Marshall BJ. Unidentified curved bacilli on gastric epithelium in active chronic gastritis. Lancet. 1983;1:1273-1275.
2) Levi S, Beardshall K, Haddad G, et al. Campylobacter pylori and duodenal ulcers: the gastrin link. Lancet. 1989;1:1167-1168.
3) Parsonnet J, Friedman GD, Vandersteen DP, et al. Helicobacter pylori infection and the risk of gastric carcinoma. N Engl J Med. 1991;325:1127-1131.
Limit: up to 100 references, 60000 characters per run
Citation authenticity workspace
Verify Citation Authenticity Before Submission
LitSource Verify compares submitted references with available DOI and scholarly metadata sources to identify verified, suspicious, and potentially fabricated entries. It is a practical citation-authenticity workspace for final checks, not a substitute for reading a source or evaluating whether it supports a claim.
What this check is for
Reference verification is most reliable when it is built into writing rather than saved for a last-minute format check. Paste the citations as they appear in the draft, keep one entry per line, and use the returned record details to identify which sources need attention. This makes it easier to catch a wrong DOI, a title that was altered during editing, or a reference copied from an unreliable draft.
LitSource Verify compares submitted references with available DOI and scholarly metadata sources to identify verified, suspicious, and potentially fabricated entries. It is a practical citation-authenticity workspace for final checks, not a substitute for reading a source or evaluating whether it supports a claim.
A repeatable verification workflow
Use the result categories to decide what to do next. A verified result means the checked metadata agrees well enough to continue with manual source review. A suspicious or fake-looking result means you should inspect the original record, correct the citation, or replace it with a source you can open. For batches, work from consequential claims outward: guidelines, clinical recommendations, efficacy statements, and safety claims deserve the earliest review.
Paste the reference exactly as it appears in your draft.
Compare identifiers and returned bibliographic fields.
Open the original source for every flagged or high-impact citation.
Example input and safe interpretation
Expected record review: compare the supplied bibliographic fields with the returned DOI or scholarly metadata record.
A strong metadata match is an authenticity signal. It does not establish claim support or replace inspection of the study.
Common citation risks to catch
Citation problems include more than nonexistent papers. A real DOI can be attached to another article, a real article can have the wrong authors or year in the bibliography, and an accurate citation can be used to support a sentence it does not justify. Those mistakes create different risks for readers and reviewers, which is why a structured verification pass is more useful than checking whether a title merely looks familiar.
What metadata matching cannot prove
Verification compares accessible bibliographic evidence. It does not establish claim support, determine study quality, confirm medical advice, or guarantee the absence of later editorial action. Source coverage differs across disciplines and publication types. A not-found result should trigger manual investigation, not an automatic conclusion that the citation is fabricated; similarly, a match should trigger source reading for important claims.
Before-submission reference control
Run verification before submission, resubmission, preprint release, editorial handoff, or publication of AI-assisted medical content. Keep a short correction log when you change an entry, particularly if the reference manager was shared with collaborators. The goal is not to make a bibliography look polished; it is to make every reference traceable enough that you can explain its origin, metadata, and relevance when a reviewer asks.
Reference-list verification example
1) Warren JR, Marshall BJ. Unidentified curved bacilli on gastric epithelium in active chronic gastritis. Lancet. 1983;1:1273-1275.
Expected record review: compare the supplied bibliographic fields with the returned DOI or scholarly metadata record.
A strong metadata match is an authenticity signal. It does not establish claim support or replace inspection of the study.
Next step: Open high-impact sources and confirm that each reference supports the sentence that cites it.