AI in Document Restoration and Manuscript Recovery Quiz

Answers and explanations appear as you take the quiz.

1. How did the Vesuvius Challenge read the carbonized Herculaneum scrolls?
   • By using CT scans and machine learning to detect ink without opening them
   • By carefully unrolling the fragile papyrus by hand
   • By soaking them in chemicals
   • By guessing the contents from the scroll's size
2. What makes detecting ink in the Herculaneum scrolls especially hard?
   • The scrolls are written in invisible ink
   • The ink and the carbonized papyrus have nearly identical density
   • The ink is bright neon and overwhelms the scan
   • There is no ink at all
3. What did DeepMind's Ithaca model do for ancient inscriptions?
   • It physically reassembled broken stone tablets
   • It translated them into modern slang only
   • It predicted missing text and estimated dates and locations
   • It painted them in color
4. What does handwritten text recognition (HTR) enable for archives?
   • Printing new copies on parchment
   • Translating speech to text
   • Colorizing the documents
   • Transcribing historical handwriting into searchable digital text
5. How do AI text-restoration models guess missing words?
   • They predict likely characters from the surrounding context
   • They copy the next document in the pile
   • They ask a librarian to fill it in
   • They always insert the most common modern word
6. What did the historical-document multispectral restoration study use multiple wavelength bands to do?
   • Help isolate degradation and reconstruct a cleaned color document image
   • Automatically read erased Herculaneum text from CT scans
   • Generate missing Greek inscriptions without source images
   • Prove every faint mark is ink
7. Why must virtual-unwrapping systems segment scroll layers before reading their surfaces?
   • To convert the scroll directly into a translation
   • To identify the ink color from its chemical formula
   • To reconstruct each curled sheet as a surface that can be flattened and inspected
   • To remove the need for X-ray imaging
8. How should scholars use a language model’s proposed restoration of missing text?
   • As proof that the missing words were present exactly as generated
   • As a substitute for the original inscription
   • As a guarantee that the inscription’s date is known
   • As candidate readings to compare with the surviving evidence