Extraction without templates
- Extracts metadata with no templates or per-document-type rules
- Multilingual: translates the descriptive vocabulary and protects titles, names and dates
- Identifies and enriches: completes the record with data the source doesn't carry, and notes the provenance of what it verifies