Paper Checker Guide¶
PaperChecker is a fact-checking system designed to validate research claims in medical abstracts by systematically searching for contradictory evidence across a document database.
Overview¶
The system performs seven key operations:
- Extracting statements from abstracts
- Generating semantic negations of claims
- Conducting multi-strategy searches
- Scoring document relevance
- Pulling supporting citations
- Synthesizing evidence reports
- Classifying findings
Verdict Classification¶
Results are categorized into three types:
| Verdict | Meaning |
|---|---|
| SUPPORTS | Counter-evidence search found no contradictions |
| CONTRADICTS | Evidence was found opposing the claim |
| UNDECIDED | Insufficient or conflicting evidence |
Search Methodology¶
PaperChecker employs three complementary search strategies:
1. Semantic Embedding Search¶
Uses vector similarity to find related documents based on meaning.
2. HyDE (Hypothetical Document Embedding)¶
Generates hypothetical documents that would contradict the claim, then searches for similar real documents.
3. Keyword Search¶
Traditional full-text search using extracted keywords.
Results are deduplicated across strategies. Higher multi-strategy matches typically indicate well-researched topics.
Usage¶
Basic Usage¶
Input Format¶
{
"abstract": "This study demonstrates that...",
"title": "Effects of Treatment X on Condition Y",
"pmid": "12345678",
"authors": "Smith J, et al.",
"year": 2024,
"journal": "Journal of Medicine",
"doi": "10.1234/example"
}
Minimum Requirements
Abstracts require minimum 50 characters of text.
Command Line Options¶
# With output file
uv run python paper_checker_cli.py abstract.json -o results.json
# Adjust scoring threshold
uv run python paper_checker_cli.py abstract.json --score-threshold 3.0
# Quick mode
uv run python paper_checker_cli.py abstract.json --quick
Confidence Assessment¶
| Level | Description |
|---|---|
| High | Strong, consistent evidence from quality sources |
| Medium | Moderate evidence with limitations |
| Low | Limited evidence or conflicting findings |
Output Format¶
{
"abstract_pmid": "12345678",
"verdict": "SUPPORTS",
"confidence": "high",
"claims_analyzed": [
{
"claim": "Treatment X reduces symptom Y",
"contradictory_evidence": [],
"supporting_evidence": [
{
"pmid": "87654321",
"title": "Confirming Study",
"relevance": 0.92
}
]
}
],
"overall_assessment": "No contradictory evidence found...",
"search_strategies_used": ["semantic", "hyde", "keyword"],
"documents_searched": 150
}
Recommended Applications¶
PaperChecker works well for:
- Validating preprint claims
- Auditing systematic reviews
- Identifying controversial findings
- Assessing training data quality
- Understanding evidence landscapes
Limitations¶
Not suited for:
- Extremely recent findings (database lag)
- Highly specialized niche topics
- Methodological statements
- Non-biomedical content
Performance Considerations¶
| Factor | Typical Value |
|---|---|
| Processing time per abstract | 2-5 minutes |
| Documents searched | 50-200 |
| Claims extracted | 3-10 per abstract |
Optimization Tips¶
- Maintain adequate database coverage
- Ensure document embeddings exist
- Adjust scoring thresholds based on desired precision vs recall
Lab Interface¶
For interactive use, the Paper Checker Lab provides a GUI:
Features:
- Abstract input with metadata fields
- Real-time progress tracking
- Interactive results exploration
- Export capabilities
Integration with Workflow¶
graph TD
A[Input Abstract] --> B[Extract Claims]
B --> C[Generate Negations]
C --> D[Multi-Strategy Search]
D --> E[Score Documents]
E --> F[Extract Citations]
F --> G[Synthesize Report]
G --> H[Classify Verdict]
Database Requirements¶
PaperChecker requires:
- PostgreSQL with pgvector extension
- Populated document database
- Pre-computed embeddings for semantic search
Best Practices¶
Abstract Selection¶
- Choose abstracts with clear, testable claims
- Ensure the topic is covered in your database
- Consider publication date relative to database currency
Interpreting Results¶
- SUPPORTS doesn't mean "proven true"
- CONTRADICTS warrants further investigation
- UNDECIDED may indicate novel research