Evaluate a large model that can be reused for several business tasks.
Consider how it can be adapted after pre-training.
Which statement best describes a foundation model?
A foundation model is broadly pre-trained and can be adapted to many downstream tasks with prompts or additional methods.
Detailed explanation
Storage is not a model.
An audit log records activity; it is not a model.
Foundation models learn broad patterns and can be adapted for different uses.
A fixed script is not a foundation model.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01ガイドのDomain 2.1で、foundation modelの定義を確認する。
Official references
What to confirm
基盤モデルを、複数用途へ適応可能な事前学習済みモデルとして説明できる。
Key points
- Pre-training
- Downstream tasks
- Adaptation
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
General capability
Pretraining provides reusable language, vision, or reasoning patterns rather than one narrow output.
Downstream task
A specific application uses or adapts the foundation model for its own inputs and acceptance criteria.