Build a chatbot that answers from frequently updated internal policies.
Use current documents rather than relying only on pre-training.
What is the primary purpose of RAG?
RAG retrieves relevant documents and adds them to the model context to ground the response in current information.
Detailed explanation
RAG retrieves context at inference time rather than retraining on every request.
RAG uses retrieved content as context for generation.
RAG uses external information to ground a generated answer.
Sensitive data must still be protected by access controls.
Check official documentation
Compare the following points with the official documentation.
Amazon Bedrock Knowledge Basesの公式ドキュメントで、検索・取得・生成の流れを確認する。
Official references
What to confirm
RAGが再学習ではなく、検索した情報による回答のグラウンディングを行うことを説明できる。
Key points
- Retrieval
- Context
- Grounding
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
Evidence freshness
The index and source documents must be updated when current information matters.
Context boundary
Retrieved content is limited by relevance, permissions, and the model's context capacity.