Does Denser AI still find the right passage when the question uses different wording?
Denser AI handled the wording mismatch and returned the supporting handbook passage. The handbook ranked first, the answer repeated its 26-day annual-leave figure with a citation, and the alternative phrase appeared only in the question.
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 is based on this test case.
Question asked with a term absent from the corpusPassEvidence
“How many days of paid time off do employees get each year?”
Full-time employees at Fenwake Instruments Ltd accrue 26 days of annual leave in each complete leave year, which runs from 1 April to 31 March.
Raw log: request, answer, and ten returned passages
Copied from Output file (text), lines 11–21 · an excerpt
Open the whole text file ↗Source excerpt · partial file · response 1
Full-time employees at Fenwake Instruments Ltd accrue 26 days of annual leave in each complete leave year, which runs from 1 April to 31 March. Part-time employees accrue this leave on a pro rata basis, and leave begins accruing from the first day of employment — it can even be taken during the probation period with the line manager's agreement.[1](upload/google_107814369447848734045/chatbots/ chatbot_o9ihun9849fcf3tbhigo4/MvdlZrB0IJ2I2OMdgJ5O9.pdf PassageID:0) Up to five days of untaken leave can be carried over into the next leave year (and must be used by 30 June), with carry-over beyond that only allowed by a director's written agreement where leave couldn't be taken due to business needs.[1](upload/google_107814369447848734045/chatbots/chatbot_o9ihun9849fcf3tbhigo4/MvdlZr B0IJ2I2OMdgJ5O9.pdf PassageID:0) --- passages (full content, all returned, none trimmed) ---
Copied from Output file (text), lines 11–21 · an excerpt
| ✓ | Expected: The supporting passage still comes back.Found: The handbook passage ranked first, the answer repeated the 26-day annual-leave figure, and the alternative phrase appeared only in the question. |
The case required the supporting passage to still come back when the question used wording the corpus never uses. The handbook passage ranked first, the answer used its 26-day figure, and the alternative phrase appeared only in the question, so a literal match was not needed.
Ranking gap. Three returned passages were from the handbook, and the handbook passage outranked the other documents; the remaining passages came from other sources and scored lower.
Citation shown. The answer included an inline citation to the uploaded handbook PDF.
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 | Managed RAG → | 15 scenarios · 9 tools |
| Capability | Retrieval → | |
| Scenario | The request uses different wording than the source → | |
| Tool | Denser AI → |
Global scenario definition → · Global capability definition → · Denser 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.