← Back to the lesson
Check whether you can judge source formats, trust, metadata, access, freshness, and synthetic data quality.
____
2. Which format often has reading-order, OCR, and table extraction risk?
4. What should happen to retired or draft documents?
6. Why is synthetic data useful for public RAG courses?
8. What is the risk of flattening JSON too aggressively?
10. Which source should usually outrank a random chat summary?