What this category's data looks like
In the compulsory education dropout prevention and control scenario, the materials are mainly of three types: dropout risk ledgers, persuasion-to-return records, and home visit records. The ledger is updated by semester or month, and its fields include student name, grade, risk level, responsible person, date, and signature; persuasion-to-return records are generated per instance and include persuasion time, participants, result, receipt, and signature; home visit records are formed per household visit and include visit date, visit target, key communication points, and parent signature. When these three types of materials enter this function, the ledger provides the risk baseline and responsible-person leads, the persuasion-to-return records provide the process and results, and the home visit records provide family communication and signature evidence. The version is based on the archived document for the current semester, with older versions retained for reference. Field alignment depends on four items: name, quantity, date, and signature; any misalignment will affect the traceability of the basis entries.
Configuration and implementation checks
This part is executed by application settings (response limit, etc.), which determine the length of a single answer and the range of context that can be accommodated. "Basis entries that can be traced back to the original text" are then completed by the model. The response limit is controlled by a slider for the length of a single answer; if set too low, the answer will be truncated, and if set too high, consumption will increase.
| Object | Check | Verification action |
|---|---|---|
| Dropout risk ledger | Fields include student name, grade, risk level, responsible person, date, and signature, and are consistent with the current semester version | In application settings, check the response limit switch and slider position, ask a basis question spanning the ledger and persuasion-to-return records, and see whether the answer gives a traceable source |
| Persuasion-to-return record | Fields include persuasion time, participants, result, receipt, and signature, and the name, quantity, and date of the same record match the ledger | Compare usage from two runs, and check whether maxResponse and quoteMaxToken match the long-answer requirement |
| Home visit record | Fields include visit date, visit target, key communication points, and parent signature, the version is current, and no signature or receipt is missing | Ask a basis question requiring a long answer and see whether the ending is complete, and record the number of citations and the maxContext range |
| Response limit | Matches the answer length required by the page, and does not control how much cited content there is | In application settings, check the switch and slider position, and confirm that longer answers have complete endings |
Three common errors
- Phenomenon: the name or date of the same student does not match between the ledger and the persuasion-to-return record. Cause: the two materials were filled in by different stages, without field alignment and version synchronization.
- Phenomenon: the home visit record lacks a parent signature or receipt, and the basis entry cannot be traced back to the original text. Cause: at archiving time, the signature column and receipt copy were not checked, and old-version materials were mixed into the current documents.
- Phenomenon: the answer breaks off in the middle of a sentence, and the basis entry is truncated. Cause: the response limit slider is set too low, so the length of a single answer is insufficient; this is unrelated to how much cited content there is.
How to verify the setup
This function's acceptance: ask a question requiring a long answer, see whether the ending is complete, and check whether maxResponse and quoteMaxToken match that length requirement.
Business test: ask a basis question spanning materials, see whether the answer gives a traceable source, and record the number of citations and the maxContext range; the two are not compared for equality.
- In application settings, check the response limit switch and slider position, confirm that they match the answer length required by the page, and compare usage from two runs.
- Ask a basis question spanning the ledger, persuasion-to-return records, and home visit records, check that the basis comes from the institution’s materials, that the version is current, and that the entries match the original text, and check whether signatures and receipts are complete.