3.3Standard
Design metrics for a summarization model.
String matching alone cannot measure factuality or readability.
Choose two appropriate practices.
× Incorrect
Use automatic metrics for broad regression checks and human review for meaning and safety, with documented criteria and agreement.
Detailed explanation
IncorrectOne person's impression is sufficient
Consistency and reviewer bias are untested.
IncorrectHigh string similarity guarantees factuality
Similar wording can still contain false claims.
CorrectUse automatic metrics for large-scale regression and humans for important samples
Scale and semantic quality are both covered.
CorrectRecord human criteria, examples, adjudication, and reviewer agreement
Subjective evaluation becomes more reproducible.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01 Domain 3.3の生成AI評価、品質指標、人手レビューを確認する。
Official references
What to confirm
自動指標と人手評価が測る品質の違いを説明できる。
Key points
- Automatic evaluation
- Human review
- Criteria
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
Automated evaluation
Using rules or metrics to compare many outputs consistently and efficiently.
Human evaluation
A structured review in which people judge outputs against defined criteria.
Question IDAWS-AIF-213
EnvironmentAWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
Last technical review2026-08-20
Report an error or rights issue