Does Dify still find the right passage when the question uses different wording from the source?
Dify retrieved the supporting passage when asked about paid time off, while the employee handbook calls it annual leave. The passage gives full-time employees 26 days per complete leave year.
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?”
Result 1 Anything beyond five days is carried over only where leave could not be taken because of a business requirement, and only with the written agreement of a director. 3.3 Leave on leaving On termination, accrued and untaken annual leave is paid in the final salary payment. Where more leave has been taken than accrued, the balance is deducted from that payment. 4. Sickness absence Report sickness to your line manager before 09:00 on the first day, by telephone rather than by message.
Result 2 Absence of seven calendar days or fewer is self-certified on return. Longer absence requires a fit note. Company sick pay is paid at normal salary for the first four weeks of a continuous absence and at half salary for the following four weeks, after which statutory sick pay alone applies. 4.1 Return to work Every absence is followed by a return-to-work conversation with the line manager. The conversation is recorded in one paragraph on the personnel file.
Result 3 A job number is required for chargeable work, for internal projects and for training. Corrections are made by the line manager rather than by editing a submitted week. 3. Annual leave Full-time employees accrue 26 days of annual leave in each complete leave year, which runs from 1 April to 31 March. Part-time employees accrue pro rata.
Result 4 Part-time employees accrue pro rata. Leave accrues from the first day of employment and may be taken during the probation period with the line manager's agreement. 3.1 Booking leave Annual leave is requested in the rota system and confirmed by the line manager. Requests are considered in the order they are received.
Show the reply as returned (Markdown)
Result 1 Anything beyond five days is carried over only where leave could not be taken because of a business requirement, and only with the written agreement of a director. 3.3 Leave on leaving On termination, accrued and untaken annual leave is paid in the final salary payment. Where more leave has been taken than accrued, the balance is deducted from that payment. 4. Sickness absence Report sickness to your line manager before 09:00 on the first day, by telephone rather than by message. Result 2 Absence of seven calendar days or fewer is self-certified on return. Longer absence requires a fit note. Company sick pay is paid at normal salary for the first four weeks of a continuous absence and at half salary for the following four weeks, after which statutory sick pay alone applies. 4.1 Return to work Every absence is followed by a return-to-work conversation with the line manager. The conversation is recorded in one paragraph on the personnel file. Result 3 A job number is required for chargeable work, for internal projects and for training. Corrections are made by the line manager rather than by editing a submitted week. 3. Annual leave Full-time employees accrue 26 days of annual leave in each complete leave year, which runs from 1 April to 31 March. Part-time employees accrue pro rata. Result 4 Part-time employees accrue pro rata. Leave accrues from the first day of employment and may be taken during the probation period with the line manager's agreement. 3.1 Booking leave Annual leave is requested in the rota system and confirmed by the line manager. Requests are considered in the order they are received.
API log: the question, four returned passages and retrieval settings
Copied from Captured file (text), lines 9–50 · an excerpt
Open the whole text file ↗Copied from Captured file (text), lines 51–90 · an excerpt
Copied from Captured file (text), lines 91–130 · an excerpt
Copied from Captured file (text), lines 131–177 · an excerpt
Source excerpt · partial file · response 1
Response body: { "query": { "content": "How many days of paid time off do employees get each year?" }, "records": [ { "segment": { "id": "434c06a2-0f8b-4be1-8612-1bdee92582e3", "position": 6, "document_id": "4c1c336c-11c5-423f-bf1e-71d339564495", "content": "Anything beyond five days is carried over only where leave could not be taken because of a business requirement, and only with the written agreement of a director.\n3.3 Leave on leaving\nOn termination, accrued and untaken annual leave is paid in the final salary payment. Where more leave has been taken than accrued, the balance is deducted from that payment.\n4. Sickness absence\nReport sickness to your line manager before 09:00 on the first day, by telephone rather than by message.", "sign_content": null, "answer": null, "word_count": 484, "tokens": 100, "keywords": [], "index_node_id": "34e1b0ec-8706-4cea-8748-e7cd73de9c0c", "index_node_hash": "9560e060d58c544a5baad0febb30576358b2a3d3f383d5ea6159aa5aef1a0a6a", "hit_count": 0, "enabled": true, "disabled_at": null, "disabled_by": null, "status": "completed", "created_by": "f1034bbd-04df-4592-ba22-56dac49ae8ad", "created_at": 1789752869, "indexing_at": 1789752891, "completed_at": 1789752896, "error": null, "stopped_at": null, "document": { "id": "4c1c336c-11c5-423f-bf1e-71d339564495", "data_source_type": "upload_file", "name": "A04_employee_handbook.docx", "doc_type": null, "doc_metadata": { "region": "group" } } }, "child_chunks": [], "score": 0.46944642,
Copied from Captured file (text), lines 9–50 · an excerpt
Source excerpt · partial file · response 2
"tsne_position": null, "files": [], "summary": null }, { "segment": { "id": "9eaf86cb-bcce-4545-8f30-d5e6cd4db0f7", "position": 7, "document_id": "4c1c336c-11c5-423f-bf1e-71d339564495", "content": "Absence of seven calendar days or fewer is self-certified on return. Longer absence requires a fit note. Company sick pay is paid at normal salary for the first four weeks of a continuous absence and at half salary for the following four weeks, after which statutory sick pay alone applies.\n4.1 Return to work\nEvery absence is followed by a return-to-work conversation with the line manager. The conversation is recorded in one paragraph on the personnel file.", "sign_content": null, "answer": null, "word_count": 460, "tokens": 91, "keywords": [], "index_node_id": "92c4f676-1613-4940-9216-b14c0bf741eb", "index_node_hash": "b3cad0103c0022456c6ea4d5203c609f2e30ff12422d2b519d80abbd73aea675", "hit_count": 0, "enabled": true, "disabled_at": null, "disabled_by": null, "status": "completed", "created_by": "f1034bbd-04df-4592-ba22-56dac49ae8ad", "created_at": 1789752869, "indexing_at": 1789752891, "completed_at": 1789752896, "error": null, "stopped_at": null, "document": { "id": "4c1c336c-11c5-423f-bf1e-71d339564495", "data_source_type": "upload_file", "name": "A04_employee_handbook.docx", "doc_type": null, "doc_metadata": { "region": "group" } } }, "child_chunks": [], "score": 0.46814263,
Copied from Captured file (text), lines 51–90 · an excerpt
Source excerpt · partial file · response 3
"tsne_position": null, "files": [], "summary": null }, { "segment": { "id": "a2399686-e1bc-453b-bcfd-ce26faa9dc08", "position": 3, "document_id": "4c1c336c-11c5-423f-bf1e-71d339564495", "content": "A job number is required for chargeable work, for internal projects and for training. Corrections are made by the line manager rather than by editing a submitted week.\n3. Annual leave\nFull-time employees accrue 26 days of annual leave in each complete leave year, which runs from 1 April to 31 March. Part-time employees accrue pro rata.", "sign_content": null, "answer": null, "word_count": 337, "tokens": 74, "keywords": [], "index_node_id": "f91121a3-e5f4-45a0-b500-abac05b64d76", "index_node_hash": "9ff9d77d07469e4b455ac1135f0867de42f38f218441ae9657ec2fee94b8b5d8", "hit_count": 0, "enabled": true, "disabled_at": null, "disabled_by": null, "status": "completed", "created_by": "f1034bbd-04df-4592-ba22-56dac49ae8ad", "created_at": 1789752869, "indexing_at": 1789752891, "completed_at": 1789752896, "error": null, "stopped_at": null, "document": { "id": "4c1c336c-11c5-423f-bf1e-71d339564495", "data_source_type": "upload_file", "name": "A04_employee_handbook.docx", "doc_type": null, "doc_metadata": { "region": "group" } } }, "child_chunks": [], "score": 0.46400315,
Copied from Captured file (text), lines 91–130 · an excerpt
Source excerpt · partial file · response 4
"tsne_position": null, "files": [], "summary": null }, { "segment": { "id": "911d0feb-05d7-4086-b9db-452d0204af85", "position": 4, "document_id": "4c1c336c-11c5-423f-bf1e-71d339564495", "content": "Part-time employees accrue pro rata. Leave accrues from the first day of employment and may be taken during the probation period with the line manager's agreement.\n3.1 Booking leave\nAnnual leave is requested in the rota system and confirmed by the line manager. Requests are considered in the order they are received.", "sign_content": null, "answer": null, "word_count": 317, "tokens": 65, "keywords": [], "index_node_id": "4cb1f7d1-a120-4483-822b-9c2a5e24c8e4", "index_node_hash": "6fa0927ae302e46f8b63c1de8306cbcbfce85250efba5e9cc96207a42fd666d6", "hit_count": 0, "enabled": true, "disabled_at": null, "disabled_by": null, "status": "completed", "created_by": "f1034bbd-04df-4592-ba22-56dac49ae8ad", "created_at": 1789752869, "indexing_at": 1789752891, "completed_at": 1789752896, "error": null, "stopped_at": null, "document": { "id": "4c1c336c-11c5-423f-bf1e-71d339564495", "data_source_type": "upload_file", "name": "A04_employee_handbook.docx", "doc_type": null, "doc_metadata": { "region": "group" } } }, "child_chunks": [], "score": 0.44551182, "tsne_position": null, "files": [], "summary": null } ] }
Copied from Captured file (text), lines 131–176 · an excerpt
| ✓ | Expected: Return the supporting passage even when the question uses different wording from the source.Found: It returned the passage giving full-time employees 26 days of annual leave per complete leave year, from 1 April to 31 March. |
The requirement was to retrieve the supporting passage, not to rank it first. Dify returned it despite the question using a term absent from the document set.
Related passages. The first, second and fourth results cover neighbouring leave and sickness-absence sections of the employee handbook.
Answer passage ranking. The passage stating the annual leave entitlement ranked third among four 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 | Managed RAG → | 15 scenarios · 9 tools |
| Capability | Retrieval → | |
| Scenario | The request uses different wording than the source → | |
| Tool | Dify → |
Global scenario definition → · Global capability definition → · Dify 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.