Does FutureSmart Document Intelligence keep corrected data downstream?
FutureSmart Document Intelligence did not keep corrected data downstream. The export path existed, but it carried only the original extraction, and the receipt row in the downloaded CSV still held RHS-70234.
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 counts this row.
Correction survives the delivery pathFailEvidence
| Object Id | Document Type | Summary | Page Number | Subtotal | Billed To | Currency | Damage Exceptions | Distance Miles | Document Outline | Entities | File Name | Fuel Rate Per Mile | Hire Period | Identifier Number | Invoice Number | Invoice Reference | Keywords | Line Items 0 Amount | Line Items 0 Description | Line Items 0 Line Total | Line Items 0 Quantity | Line Items 0 Unit Price | Line Items 1 Amount | Line Items 1 Description | Line Items 1 Line Total | Line Items 1 Quantity | Line Items 1 Unit Price | Line Items Remaining Amount | Line Items Remaining Description | Line Items Remaining Line Total | Line Items Remaining Quantity | Line Items Remaining Unit Price | Order Reference | Organizations | Receipt Number | Received By | Reference Number | Site | Tax | Temporal Context | Title | Topics | Total Due | Total Fuel Cost | Vehicle Registration |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 4daa8753-c6f9-49a9-bfaf-03c9be8656ed | invoice | This invoice from Thistlewood Plant & Tool Hire to Oakbridge Construction Ltd covers the weekly hire of site equipment including a telehandler, generators, and site fencing. The total amount due for the period of 03 March to 10 March 2026 is £1,754.40. | 1462 | Oakbridge Construction Ltd | GBP | Header with vendor and client details, hire period, line items table, subtotal, tax, total due, and remittance instructions. | Thistlewood Plant & Tool Hire; Oakbridge Construction Ltd | SiblingA1_Thistlewood_PlantHire.pdf | 03 Mar - 10 Mar 2026 | TPH-3301 | Telehandler; Generator; Site fencing | Telehandler, 4-tonne, weekly hire, 1 unit at 850.00 each, total 850.00 | Generator, 20kVA, weekly hire, 2 units at 210.00 each, total 420.00 | Site fencing, per 10m panel, weekly, 8 units at 24.00 each, total 192.00 | Thistlewood Plant & Tool Hire; Oakbridge Construction Ltd | Oakbridge Depot Extension | 292.4 | 03 Mar - 10 Mar 2026 | Invoice TPH-3301 | plant hire; construction equipment | 1754.4 | ||||||||||||||||||||||||
| 75285524-15d3-4d21-9000-cf4c1bc8e326 | invoice | This invoice from Meridian Retail Group to Blackwood Distribution LLC details the supply of various retail fixtures and store equipment. The document lists ten line items including shelving, checkout counters, and scanning equipment, totaling £6,591.24 including VAT. | main_invoice_multipage_blackwood.pdf | 1008 | Retail Shelving Unit - Steel, 5-Tier | 12 | 84 | 930 | Checkout Counter - Laminate Top | 3 | 310 | 273; 648; 304; 1080; 580; 330; 570; 380 | Security Tag Deactivator - Handheld; Price Label Printer - Thermal; Shopping Basket - 20L, Set of 10; Storeroom Pallet Racking - 4-Bay; Barcode Scanner - Wireless; Receipt Paper Roll - Case of 50; Mannequin - Adjustable, Full Body; Display Signage Kit - Window Set | 6; 4; 8; 2; 10; 15; 6; 5 | 45.5; 162; 38; 540; 58; 22; 95; 76 | ||||||||||||||||||||||||||||||
| cbb19486-9090-445b-a340-e940905e9bb1 | receipt | This document is a service receipt from Riverside Home Services for a boiler annual inspection performed for Mr. Grant Sullivan. It details the service provided, the subtotal, the applied tax, and the final total amount due. | 85 | USD | TC1_TC4_TC5A_baseline_receipt_1.pdf | RHS-70234 | $5.10 | 90.1 | |||||||||||||||||||||||||||||||||||||
| dd1200e5-4695-4d43-84a5-9077b688d3e7 | invoice | This invoice from Calderwood Freight Services details the road haulage charges for a consignment transported from Leeds to Portsmouth. The document confirms the successful delivery of goods on March 7, 2026, with no reported damages or exceptions. | GBP | 450 | Calderwood_Freight_S66_S67_S72.pdf | 0 | 81 | ||||||||||||||||||||||||||||||||||||||
| 45a86024-1893-4ef0-b83e-b65858eebb1d | invoice | This document is an invoice issued by Meridian Retail Group to Blackwood Distribution LLC for the supply of various retail fixtures and store equipment. It details ten line items, including shelving, checkout counters, and scanning equipment, with a total amount due of £6,591.24. | 2 | £6,103.00 | main_invoice_multipage_blackwood.pdf |
CSV export: 17 collection rows
Copied from Output file (CSV), lines 1–6 · an excerpt
Open the whole CSV file ↗| ✗ | Expected: The delivery path carries the corrected value.Found: The export exists, but it carries the original extraction rather than a corrected value. |
The tool’s response is printed above, under What the tool returned. Show this screenshot
Open original ↗The tool’s response is printed above, under What the tool returned. Show this screenshot
Open original ↗The export path exists, but the product cannot change a value before exporting it, so the downstream copy still shows the original extraction. That makes this a Fail rather than a pass.
Export wrote file. The Export CSV action wrote a CSV file, and the receipt value appears in the exported receipt row.
Rows per document. The export is collection-level: it produced one row per document across 17 documents and nine source PDFs.
Schema flattening. The CSV had 46 columns, matching the 32-column schema after the line-items group was flattened.
Second path exists. The product also documents an API delivery path.
The export covers the whole collection and does not show a corrected value or edit history.
The toast alone does not show a file on disk; the Downloads view and opened CSV do, and no edit, correction, or re-export control is shown.
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 | Review and Correction → | |
| Scenario | Corrected data goes downstream → | |
| Tool | FutureSmart Document Intelligence → |
Global scenario definition → · Global capability definition →
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.