Paper Reference Checker#

Systematically verify academic references to detect AI-hallucinated or fabricated citations. Queries Google Scholar, arXiv, CNKI, and other databases.

---

Core Workflow#

Phase 1: Citation Extraction (Token-Efficient First)#

ALWAYS use targeted extraction before full-document reading — saves 80–95% tokens.

Input TypePrimary MethodFallbackarXiv linkarxiv.org/html/{ID} → find references sectionFull HTML, then PDFPDF fileLast 15–20% of pages onlyExpand to 30% → 50% → fullOverleaf linkRegex cite-keys from .tex → filter .bib/.bblInline bibitem in .texPasted listParse directly—

Phase 2: Multi-Platform Querying (Priority Order)#

1. DOI → https://doi.org/{DOI} — resolves = ✅ confirmed
2. arXiv ID → https://arxiv.org/abs/{ID} — match = ✅ confirmed
3. Google Scholar → search "Full Title"
4. arXiv search → arxiv.org/search/
5. CNKI → cnki.net
6. Fallbacks: Semantic Scholar · PubMed · IEEE Xplore · ACM DL · DBLP

Phase 3: Authenticity Judgment#

StatusLabelCriteria✅VERIFIEDFound in ≥1 authoritative DB⚠️UNCERTAINPartial match❌NOT FOUNDNo match across all queried channels🔴FABRICATEDNon-existent venue, unresolvable DOI🔗BROKEN CITATION[?] marker in PDF body

Phase 4: Report Output#

See examples/sample-report.md for full example.

Support Files#

FilePurposereferences/citation-extraction.mdFormat rulesreferences/search-strategies.mdPer-database query tacticsreferences/verification-criteria.mdDecision flowchartscripts/extract_references.mdFull decision treeexamples/sample-report.mdComplete report exampleexamples/bibtex-example.bibAnnotated BibTeX

Databases#

Google Scholar · arXiv · CNKI · Semantic Scholar · PubMed · IEEE Xplore · ACM DL · DBLP · DOI Resolver · Crossref