A general text model is adapted to classify medical documents.
Use the terms precisely in the design document.
Which statement is correct?
Pre-training learns broad capabilities from large data, while fine-tuning adds task- or domain-specific training.
Detailed explanation
Pre-training is a prior training stage, not automatic per-request updating.
A customized model still requires quality, bias, and safety evaluation.
A model can learn general patterns first and then update parameters using task data.
Changing a prompt is prompt engineering; fine-tuning is additional model training.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01ガイドのDomain 2.2で基盤モデルの学習・適応用語を確認する。
Official references
What to confirm
事前学習、微調整、プロンプト変更の違いを説明できる。
Key points
- General capability
- Additional training
- Continued evaluation
Notes
- Environment: AWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
- Compare the definition with the AIF-C01 objective and the service documentation before applying it.
- Keep the concept separate from a particular answer choice or implementation step.
Foundation review
Parameter change
Fine-tuning changes learned parameters and therefore requires data quality, evaluation, and release controls.
Knowledge update
Changing source material may be handled through retrieval without retraining model parameters.