Nested & Repeated Fields
Returns repeated sets of records completely, with each value under the right record: none dropped, none invented, none cross-bound. ON TRIAL: merges back into Field Extraction if precision and recall do not actually diverge on repeated sets.
What this capability means
A tool has this capability when it can take a schema for repeated records and return every genuine member once, with each value attached to the correct record. It must not invent members, drop members, or mix values between records.
Boundary: It does not judge single-field extraction on its own.
Scenarios that test this capability
A scenario is a real-world situation used to test a capability.
A repeated set of records
Scenario
The set contains a non-record
Scenario
The set is legitimately empty
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?