Can Landing AI answer a document question that isn't in the extraction schema?
Landing AI answered a document question outside the extraction schema and tied the answer to the source document. In the invoice test, it returned 06 Jul 2020 for the due date and showed a visual reference to the row containing that value. A second wording returned the same attributed answer.
1 of 1 test case passed
Every test case in this scenario has a result.
The test case
Each test case is judged on its own: Pass, Fail, Not gradable, or Untested. The scenario result above counts this row.
Answer outside the schema, attributedPassEvidence
Input 1: “What is the due date on this invoice?”
Input 2: “By which date does this invoice have to be paid?”
The tool's reply is shown in the screenshots under Proofs.
| ✓ | Expected: It distinguishes between extracted data and the document, either saying the field is absent or answering from the document with attribution.Found: It answered from the document, returned 06 Jul 2020, and showed a visual reference to the invoice row containing that date. |
{ "surface": "Tool > Chat (labelled Experimental; works with this file only, not available as an API)", "file": "main_invoice_multipage_blackwood.pdf (in the 3-file project)", "schema_in_project": [ "invoice_reference", "total_due", "total_due_all_documents" ], "turns": [ { "question": "What is the due date on this invoice?", "answer": "The due date on the invoice is 06 Jul 2020.", "visual_reference": "5. table" }, { "question": "By which date does this invoice have to be paid?", "answer": "The invoice has to be paid by 06 Jul 2020.", "visual_reference": "5. table" } ], "note": "The first question was asked three times in total across the day; the answer was the same each time." }
It checked whether the answer was treated as schema data or tied to the document. The product returned 06 Jul 2020 and showed a visual reference to the invoice row with that date; a second wording produced the same attributed answer.
Chat limitation note. The chat view says it works with this file only and is not available as an API.
Configuration and setup
How this tool was set up for the run and what the test needed in place. Each row is a fact from the run's records; a fact the records do not hold is left out, not guessed.
How this scenario is graded
How we decide Pass, Fail and Not gradable. The same rules apply to every tool tested on this scenario.
Each test case gets one result per tool: Pass, Fail or Not gradable. A test case we haven't run yet shows Untested. There are no partial results.
- Pass: the tool did everything the test expected, and nothing it said contradicts the correct answer.
- Fail: at least one expected behaviour clearly didn't happen; the row says which and quotes the tool.
- Not gradable: our evidence couldn't settle the outcome (for example a record we needed is missing). It is never counted as a fail, and the row says what's missing.
Where this sits in the benchmark
This page is one cell of a larger study: one tool, one scenario. Only this benchmark's frame appears here.
| Level | Name | Scope |
|---|---|---|
| Benchmark | Structured Document Extraction → | 19 scenarios · 10 tools |
| Capability | Querying → | |
| Scenario | The answer was never in the schema → | |
| Tool | Landing AI → |
Global scenario definition → · Global capability definition → · Landing AI product page →
History of this result
What has happened to this result since it was first published. Runs and grades are never overwritten: a retest or a re-grade publishes a new result and keeps the earlier one readable.
Act on this result
Nothing filed here edits the run or the grade. A challenge opens a review, and a review can produce a new run or a re-grade — which becomes the current result and leaves this one in the history.
You run the same kind of test against your own setup and get the same behaviour.
Agree →Yours behaves differently. Tell us what you got, with a screenshot if you have one.
Disagree →Something here is wrong — a reference value, a transcription, a grade.
Report an issue →Tell us what changed and we schedule a rerun of the failing test case. The old result stays as history.
Vendor notice →The same record is available as structured data through the AI Demos MCP server, with the counts, the coverage and every per-test-case reason carried as fields.
Verify the proof files
These files support this result. Open a file to inspect the original evidence.
File fingerprints (SHA-256)
A fingerprint identifies the exact file used for this result.