What this category's data looks like
This page focuses on the grade entry and credit recognition scenario. Material sources include transcripts, deferred examination applications, and makeup examination arrangements, and the update cadence rolls with the academic affairs milestones of each semester. Transcripts are the core evidence for credit recognition, deferred examination applications are the trigger documents for temporarily missing grades and subsequent entry, and makeup examination arrangements are the basis documents for grade replacement and credit supplementary entry. When the three types of materials enter this feature, each is stored as an independent file record and data row identity, with field characteristics concentrated in course name, credit count, examination date, signature, and receipt. If the name, count, or date of the same record does not match across two materials, the version is not synchronized, or the signature or receipt is missing, the basis entry cannot be traced back to the original text. After material parsing and chunking, data is immediately stored as visible data rows for subsequent retrieval and business processing, and cross-checking between materials relies on basis entries that can be traced back to the original text.
Configuration and implementation checks
This step is executed by the parsing and training task (server-side task), which outputs data rows and index status (parsed, indexing, indexed). "Basis entries that can be traced back to the original text" are then completed by retrieval and subsequent processing. After parsing is complete, parsing and chunking immediately store data as visible data rows and bind a unique data identifier. The index transitions through the three states parsed, indexing, and indexed. Data pending indexing does not participate in recall, and changing the model only rebuilds data that has already completed indexing.
| Object | Check | Verification action |
|---|---|---|
| Transcript | Whether the course name, credit count, examination date, and signature match the deferred examination application and makeup examination arrangement | Open the data row to check whether the status is parsed, indexing, or indexed, then compare the course name and credit count field by field |
| Deferred examination application | Whether the deferred examination course, application date, and signature receipt are consistent with the transcript and makeup examination arrangement | Check the file record and the data row separately, and confirm that the version is current and the signature receipt is complete |
| Makeup examination arrangement | Whether the makeup examination course, makeup examination date, and grade replacement relationship are consistent with the transcript and deferred examination application | Check whether the data rows covered by the rebuild task contain only data that has completed indexing, then compare the date and course name |
| Parsing and training task | Whether the data row status is consistent with the interface display, and whether the file record and data row identities are distinguishable | Open the data row to check the three states, and check chunkSize, chunkSplitter, trainingType, and the index status |
Three common errors
- Symptom: a newly uploaded material cannot be found. Cause: the data row is still in parsed or indexing, data pending indexing does not participate in recall, and successful parsing does not equal a retrieval hit.
- Symptom: the name, count, or date of the same course does not match across two materials. Cause: the material versions are not synchronized, or the signature and receipt are missing, so the basis entry cannot be traced back to the original text.
- Symptom: empty data rows are left after a training failure. Cause: residual data from an interrupted training is not cleaned up, or after changing the model the rebuild scope mistakenly covers data pending indexing.
How to verify the setup
Feature acceptance: upload one material, observe the data row status transitions and the coverage scope of the rebuild task, and confirm that the status passes through parsed, indexing, and indexed in sequence, and that the rebuild covers only data that has completed indexing.
Business test: ask a cross-material basis question, see whether the answer provides a traceable source, and check that the citation count and citation text tokens are recorded separately.
- Open the data row to check whether the status is parsed, indexing, or indexed, and confirm that it is consistent with the interface display.
- Check the file record and the data row separately, and confirm that the rebuild task covers only data that has completed indexing, and that data pending indexing does not participate in the rebuild.