Can Reducto repair a record by adding one member and removing one?
Reducto could not repair the extracted set: its Studio results table has no add-row or delete-row controls. Reading the set back still showed the original three items: Site fencing remained and the Concrete mixer was never added.
1 of 1 test case failed
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 is based on this test case.
Add one member and remove oneFailEvidence
On the extracted invoice in Reducto Studio, we set out to remove the Site fencing (per 10m panel, weekly) row and add a new Concrete mixer (150L, weekly hire) row with quantity 1, unit price £95.00 and amount £95.00.
{
"instructions": {
"schema": {
"type": "object",
"properties": {
"line_items": {
"type": "array",
"items": {
"type": "object",
"properties": {
"description": { "type": "string" },
"quantity": { "type": "string" },
"unit_price": { "type": "string" },
"amount": { "type": "string" }
},
"required": ["description", "quantity", "unit_price", "amount"]
}
}
},
"required": ["line_items"]
}
},
"settings": {
"citations": { "enabled": true }
}
}{ "line_items": [ { "description": { "value": "Telehandler, 4-tonne, weekly hire", "citations": [ { "type": "Text", "bbox": "{… 6 fields in the file}", "content": "Telehandler, 4-tonne, weekly hire", "image_url": null, "chart_data": null, "confidence": "high", "granular_confidence": "{… 2 fields in the file}", "extra": null, "parentBlock": "{… 8 fields in the file}" } ] }, "quantity": { "value": "1", "citations": [ { "type": "Text", "bbox": "{… 6 fields in the file}", "content": "1", "image_url": null, "chart_data": null, "confidence": "high", "granular_confidence": "{… 2 fields in the file}", "extra": null, "parentBlock": "{… 8 fields in the file}" } ] }, "unit_price": { "value": "£850.00", "citations": [ { "type": "Text", "bbox": "{… 6 fields in the file}", "content": "£850.00", "image_url": null, "chart_data": null, "confidence": "high", "granular_confidence": "{… 2 fields in the file}", "extra": null, "parentBlock": "{… 8 fields in the file}" } ] }, "amount": { "value": "£850.00", "citations": [ { "type": "Text", "bbox": "{… 6 fields in the file}", "content": "£850.00", "image_url": null, "chart_data": null, "confidence": "high", "granular_confidence": "{… 2 fields in the file}", "extra": null, "parentBlock": "{… 8 fields in the file}" } ] } }, { "description": { "value": "Generator, 20kVA, weekly hire", "citations": [ { "type": "Text", "bbox": "{… 6 fields in the file}", "content": "Generator, 20kVA, weekly hire", "image_url": null, "chart_data": null, "confidence": "high", "granular_confidence": "{… 2 fields in the file}", "extra": null, "parentBlock": "{… 8 fields in the file}" } ] }, "quantity": { "value": "2", "citations": [ { "type": "Text", "bbox": "{… 6 fields in the file}", "content": "2", "image_url": null, "chart_data": null, "confidence": "high", "granular_confidence": "{… 2 fields in the file}", "extra": null, "parentBlock": "{… 8 fields in the file}" } ] }, "unit_price": { "value": "£210.00", "citations": [ { "type": "Text", "bbox": "{… 6 fields in the file}", "content": "£210.00", "image_url": null, "chart_data": null, "confidence": "high", "granular_confidence": "{… 2 fields in the file}", "extra": null, "parentBlock": "{… 8 fields in the file}" } ] }, "amount": { "value": "£420.00", "citations": [ { "type": "Text", "bbox": "{… 6 fields in the file}", "content": "£420.00", "image_url": null, "chart_data": null, "confidence": "high", "granular_confidence": "{… 2 fields in the file}", "extra": null, "parentBlock": "{… 8 fields in the file}" } ] } }, { "description": { "value": "Site fencing, per 10m panel, weekly", "citations": [ { "type": "Text", "bbox": "{… 6 fields in the file}", "content": "Site fencing, per 10m panel, weekly", "image_url": null, "chart_data": null, "confidence": "high", "granular_confidence": "{… 2 fields in the file}", "extra": null, "parentBlock": "{… 8 fields in the file}" } ] }, "quantity": { "value": "8", "citations": [ { "type": "Text", "bbox": "{… 6 fields in the file}", "content": "8", "image_url": null, "chart_data": null, "confidence": "high", "granular_confidence": "{… 2 fields in the file}", "extra": null, "parentBlock": "{… 8 fields in the file}" } ] }, "unit_price": { "value": "£24.00", "citations": [ { "type": "Text", "bbox": "{… 6 fields in the file}", "content": "£24.00", "image_url": null, "chart_data": null, "confidence": "high", "granular_confidence": "{… 2 fields in the file}", "extra": null, "parentBlock": "{… 8 fields in the file}" } ] }, "amount": { "value": "£192.00", "citations": [ { "type": "Text", "bbox": "{… 6 fields in the file}", "content": "£192.00", "image_url": null, "chart_data": null, "confidence": "high", "granular_confidence": "{… 2 fields in the file}", "extra": null, "parentBlock": "{… 8 fields in the file}" } ] } } ] }
Stored job record: the same three line items read back after the attempts
Copied from Output file (JSON), result · an excerpt
Open the whole JSON file ↗| ✗ | Expected: After one item is added and one removed, the stored set has the addition, lacks the removed item, and every other item is untouched.Found: The stored set still held the same three items, Telehandler, Generator and Site fencing; no Concrete mixer item existed, and no other item's values changed. |
- job_id
- 2ad93760-d78d-4a7e-8538-950c35dde3e0
It is a Fail because the product lacks set-level editing, not because the run went wrong: the extraction returned the three items correctly, but the results table offers no way to add or delete a row, and Reducto's Edit feature changes the source document, not the results.
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 test case's result says which one 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 test case's result 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 | Review and Correction → | |
| Scenario | Repair a record → | |
| Tool | Reducto → |
Global scenario definition → · Global capability definition → · Reducto 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.