What this category's data looks like
In the compulsory education dropout prevention and control scenario, the materials entering knowledge base retrieval are mainly of three types: dropout risk ledgers, persuasion-to-return records, and home visit records. Risk ledgers usually contain student identifiers, risk levels, determination dates, and handler signatures, and serve as the base register for determining whereabouts and authorization scope; persuasion-to-return records contain persuasion time, participants, results, and receipts, and are used to verify the dates and quantities of the same event; home visit records contain home visit dates, visitors, and parent signatures or receipts, and are used to supplement process evidence. When the three types of materials enter the authorized access and external service function, each plays a different role: the ledger determines the visible scope and identity matching, while persuasion-to-return records and home visit records provide source text fragments for traceability. Materials are updated by semester and by month, and field names, date formats, and signature fields are prone to inconsistency across different materials, so the source text fragment back-reference must be treated as authoritative.
Configuration and implementation checks
This step is executed by the citation output of the knowledge base retrieval node, which outputs an array of cited content, each item containing a source text fragment and a source identifier, and the length may be 0. The externally available entry point and authorization scope are then completed by subsequent AI nodes, code nodes, or manual work.
| Object | Check | Verification action |
|---|---|---|
| Dropout risk ledger | Whether the fields, dates, and signatures are consistent with the persuasion-to-return records and home visit records | In the node, check the selected collection and its vector model, compare the risk level, determination date, and handler signature one by one, and record the length of the citation array |
| Persuasion-to-return record | Whether the name, quantity, and date of the same record match across the two materials | Open the citation marker to back-reference the source paragraph, verify the persuasion time, participants, results, and receipt, and record empty arrays as normal output as well |
| Home visit record | Whether the version is synchronized, and whether the signature or receipt is missing | Check the home visit date, visitor, and parent signature against the source document, use quoteList to view the matched materials, and use responseData (non-streaming module results returned with detailed output enabled) to record the output chain |
| Search parameters | Whether they are consistent with the values set in the knowledge base retrieval section | Open the search parameters panel and check item by item; verify the tokens of the cited text against the quoteMaxToken budget, and record the number of array items separately |
Three common errors
- Symptom: the name, quantity, and date of the same persuasion-to-return record do not match between the ledger and the home visit materials. Cause: the two materials were entered separately and did not back-reference the same source via the source text fragment.
- Symptom: the home visit record version is not synchronized, and the old version's signature field is missing. Cause: after the material was updated, it was not re-entered into the knowledge base, and the citation still points to the old paragraph.
- Symptom: the answer has content but the citation array is empty, or the cited fragment does not match the source text. Cause: no material was matched in this round, or a different item was matched; the match must be checked first before checking trimming.
How to verify the setup
This function's acceptance: ask a question with source-text support and check whether the citation array and markers can back-reference the source text; then ask a question with no matching material and confirm that the array is empty while the chain still runs through.
Business test: access once with a restricted identity and see whether the visible scope matches the setting, and whether unauthorized content is invisible.
- Run a test with a query that has no matching material, see that the output is empty while the chain continues, and record the length of the citation array.
- Open the citation markers in the conversation and check against the source document whether each citation can back-reference the source paragraph.