Compare two summarization models.
String-overlap metrics alone do not fully measure readability or factuality.
Choose two useful human-evaluation criteria.
Human review can assess whether key facts are preserved and whether the output is safe and appropriate.
Detailed explanation
Billing information does not measure summary quality or safety.
Human review can identify safety and policy violations.
A name ordering is not a quality metric.
Reviewers can assess faithfulness and missing key information.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01ガイドのDomain 3.3で人手評価と自動評価の役割を確認する。
Official references
What to confirm
自動指標で測りにくい意味・安全性の評価項目を挙げられる。
Key points
- Faithfulness
- Safety
- Human 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
Semantic judgment
A fluent response can still omit important facts or introduce unsupported claims.
Reviewer agreement
Clear criteria and examples help different reviewers apply the same standard.