This website uses essential cookies for authentication, language preferences and remembering that you dismissed this notice. No tracking or advertising cookies are used. Privacy policy

Identify supported publications, metadata discrepancies and references needing editorial review.

Hallucite gathers available, applicable publication evidence for reference review, with metadata discrepancies and unresolved cases visible to your editorial team.

Contact Hallucite

04 / Identifier integrity

A working DOI still needs a metadata comparison

A DOI can resolve to a real publication while a reference gives the wrong year or other details. Compare the identifier record with the cited fields and investigate the differences.

Hallucite checks available publication evidence. A matching identifier does not establish correct metadata, authorship, intent or support for a claim.

Illustrative reference comparisonDOI 10.1093/mind/LIX.236.433

A real publication and a deliberately altered reference, not a checker result or an accuracy test. Oxford Academic publisher record

Publisher reference

Turing, A. M. (1950). Computing machinery and intelligence. Mind, 59(236), 433–460.

The publisher dates this article to 1950. The DOI identifies that work.

Year in publisher record: 1950

Deliberately altered reference

Turing, A. M. (2020). Computing machinery and intelligence. Mind, 59(236), 433–460.

This teaching example keeps the same DOI and changes only the cited year to 2020.

Year comparison: cited 2020 · record 1950 — review the difference

What we check — and what we don't

Every verdict is limited to what the evidence supports. Hallucite verifies the reference record without making claims about authorship or intent.

What Hallucite checks

  • Publication matches: available records support the cited work.
  • Identifier integrity: compare the cited DOI or PMID with the publication it resolves to.
  • Metadata consistency: compare authors, year, journal and title.
  • Publication notices: show available retraction and correction evidence.

What we don't do

  • We don't score whether your text was written by AI.
  • We don't yet judge whether a source supports the specific claim it's attached to.

Verification sources

  • Crossref
  • PubMed
  • NLM Catalog
  • OpenAlex
  • CiNii
  • DBLP
  • Open Library
  • Google Books
  • Semantic Scholar
  • DOI registry

Built for editorial workflows

  • 1

    Auditable evidence per reference

    Results show publication matches, metadata discrepancies and unresolved entries, with source evidence and available publication notices for your editorial review.

  • 2

    Batch processing and API access

    Submit reference lists through the web interface or REST API. Export the results for review in your editorial workflow.

  • 3

    Exports for your editorial record

    Use the available structured results in your editorial review. Keep the reference, source links, field differences and unresolved questions together in the editorial record.

How we measure and what remains uncertain

Hallucite makes deterministic, versioned verification decisions and shows the evidence for each flag. Hosted in the EU.

Coverage gaps, incomplete metadata and source outages can leave a reference unresolved. Review flagged entries: a missing match does not prove fabrication, and a verdict does not establish authorship or whether a source supports a claim.

Benchmark results are still to come and will be published after review.

Read the methodology

Questions about reference checking?

Contact us with the workflow or service question you want to understand. Use the pricing page for current plans and limits.

Contact Hallucite