OCR
The same schema returns the same values when the document is an image, and a mixed document is handled page by page.
What this capability means
OCR means a tool can return the same structured-record values from scanned or image-only documents as it does from digital ones, and can handle documents that mix digital and scanned pages page by page. The judged output is the returned record, not the transcription.
Boundary: It does not score transcription quality.
Scenarios that test this capability
A scenario is a real-world situation used to test a capability.
A cleanly scanned document
Scenario
Benchmarks that include this capability
A capability has global identity and may be used by more than one benchmark.
Structured Document Extraction
Which document extraction platform turns a user-defined schema into correct structured records — across layouts, scans and repeated sets — and is honest about what it could not find?