What this category's data looks like
This page addresses the scenario of annual progress and acceptance of research projects; the materials include annual progress reports, mid-term review materials, and acceptance reports. Annual progress reports focus on phase tasks, indicator completion status, and fund usage; mid-term review materials focus on milestone results, issues, and adjustment explanations; acceptance reports focus on final indicators, conclusions, and signed receipts. The three types of materials often contain multilingual materials and terminology. When entering the model access and configuration stage, they serve as input sources for dialogue and retrieval respectively: progress reports are used to extract terminology and reference drafts, mid-term review materials are used to verify how terminology is written in process documents, and acceptance reports are used to check final translated names and back-references to the source text. Material updates roll forward with review milestones; fields involve names, quantities, dates, and signatures, and version synchronization must be maintained.
Configuration and implementation checks
This stage is executed by the server-side model channel configuration, outputting model binding relationships and channel responses. The unified terminology table and reference drafts are completed subsequently by AI nodes, code nodes, or manual work. Models are divided into five types: language, indexing, reranking, speech synthesis, and speech recognition. Different stages bind different model types; during configuration, the type labels and the model names returned by the channel must be checked.
| Object | Check | Verification action |
|---|---|---|
| Annual progress report | The names, quantities, and dates of the same record in the progress report and mid-term review materials must correspond | Check the report fields against the mid-term material fields, check whether terminology writing is consistent throughout, and compare against the model names returned by the channel |
| Mid-term review materials | The version must be synchronized with the annual progress report; missing signatures or receipts must be completed | Check the version number and signed receipts, check for indexing model-related errors, and confirm that the maxContext and maxTokens settings match the material scale |
| Acceptance report | The final indicators, conclusions, and signatures must be consistent with earlier materials | Check the names, quantities, and dates in the acceptance report and progress report, check whether translated names can back-reference the source text, and confirm the quoteMaxToken and toolChoice configurations |
| Model type | Corresponds to the stage in which it is to be used | Check the type labels in the model configuration, distinguishing language, indexing, reranking, speech synthesis, and speech recognition |
Three common errors
- Symptom: The names, quantities, and dates of the same record in the annual progress report and mid-term review materials do not match. Cause: The two materials are maintained separately, without field-level comparison and version synchronization.
- Symptom: The acceptance report lacks signatures or receipts, blocking external release. Cause: The signature process is not bound to the material version, and receipts are not updated with the version.
- Symptom: Retrieval reports indexing model-related errors, or the answer style changes abruptly after switching models. Cause: The model type does not match the stage, or the model identifier is inconsistent with the channel response.
How to verify the setup
Acceptance of this feature: run one dialogue and one retrieval, check the model identifier in the response, confirm it is consistent with the names listed in the document, and check the configurations related to maxContext, maxTokens, quoteMaxToken, and toolChoice.
Business test: sample three terms and check whether their writing is consistent throughout, check whether translated names can back-reference the source text, and record the tokens and array count of the quoted text.
- Check the channel configuration and key source, confirm they are consistent with the connected service, and compare against the model names returned by the channel.
- Check the type labels in the model configuration, and confirm that language, indexing, reranking, speech synthesis, and speech recognition are bound to their corresponding stages respectively.