Scenario in benchmark · Version 1
Correct a field value
A human-applied correction changes a returned field value in the stored record.
How the tools performed
Every in-scope tool is visible. Outcomes come from current published Results for this scenario and benchmark version.
No published results for this scenario yet.
No published result does not tell you whether a tool has been tested. Test coverage is shown only for comparable published Results.
| Tool | Published outcomes | Test coverage | Result |
|---|---|---|---|
| No published result · alphabetical | |||
| Datalab | — | — | No published result |
| Docsumo | — | — | No published result |
| Extend AI | — | — | No published result |
| FutureSmart Document Intelligence | — | — | No published result |
| Landing AI | — | — | No published result |
| LlamaParse | — | — | No published result |
| Nanonets | — | — | No published result |
| Reducto | — | — | No published result |
| Retab | — | — | No published result |
| Unstract | — | — | No published result |
Assessed = Pass + Fail + Not gradable. Gradable = Pass + Fail. Both use the published Result’s pinned-test denominator. — means not publicly available.
Test design
- Pinned test cases
- 1
- Disclosure
- 0 public · 1 withheld
- Capabilities exercised here
- Review and Correction
What this scenario evaluates
- Whether a human-applied change becomes the stored field value.
- Whether the corrected value is still present when the record is read again.
- Whether the original returned value is replaced rather than kept as the stored value.
Exact wording, inputs, fixture state, expected output and detailed grading remain at the test-case level and may be withheld while the benchmark version is active. The scenario and its evaluation intent are public.
How the results are graded
- Pass: the test-case expectations hold.
- Fail: an expectation demonstrably does not hold.
- Not gradable: the evidence cannot establish the outcome.
Version 1 uses test-case expectations; no scenario rubric is pinned.
Benchmark methodology →