Can FutureSmart Document Intelligence handle requests that use wording different from the source?
FutureSmart Document Intelligence handled the wording change and still returned the annual-leave answer. Search ranked the handbook passage first, and chat answered 26 days per complete leave year with part-time pro rata.
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.
Question asked with a term absent from the corpusPassEvidence
Input 1: “How many days of paid time off do employees get each year?”
Input 2: “How many days of paid time off do employees get each year?”
Response 1, the final response, answers Input 2.
Full-time employees receive **26 days of annual leave per complete leave year**. Part-time employees receive leave on a pro-rata basis.
Text record: the search and chat calls and their outputs
Copied from Output file (text), lines 92–94 · an excerpt
Open the whole text file ↗Source excerpt · partial file · response 1
status: 200 answer: Full-time employees receive **26 days of annual leave per complete leave year**. Part-time employees receive leave on a pro-rata basis.
Copied from Output file (text), lines 92–94 · an excerpt
| ✓ | Expected: The supporting passage still comes back.Found: Search returned the handbook's annual-leave passage first, and chat answered 26 days per complete leave year with part-time pro rata. |
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 request uses different wording, and the handbook uses annual leave instead. Search ranked that passage first, and chat answered 26 days per complete leave year with part-time pro rata.
Shared terms. The top chunk shared days, employees, year, paid, time and each with the question; 'paid time off' and 'PTO' were absent from the documents.
Section count. The chat call returned 27 sections, and most of the rest scored below 0.33.
One question was run, so this shows one paraphrase pair, not general synonym handling.
The broad top chunk shows the passage came back, not how tightly it is isolated.
The notebook view is truncated; the attached text file has the rest.
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 | Managed RAG → | 15 scenarios · 8 tools |
| Capability | Retrieval → | |
| Scenario | The request uses different wording than the source → | |
| 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.