MP-DocVQA

Questions over multi-page scanned industry documents.

The evaluated unit is a model response or component output. Read the source for the exact prompt, tool and harness conditions.

Answer exact match and page retrieval accuracy.

Scores remain in the original unit. They are not normalised or combined with results from other benchmarks.

• Primary benchmark source

Relevant to reading scanned legal documents, though the benchmark itself is not law-specific.