Tune output diversity.
Do not treat sampling settings as a factuality guarantee.
Choose two correct statements.
Top-p restricts sampling to a cumulative-probability set and temperature changes sampling randomness; both should be measured on representative inputs.
Detailed explanation
The effect depends on model and task.
Sampling alone cannot provide evidence or correctness.
It changes the candidate set and can affect diversity.
High diversity can make format and safety less stable.
Check official documentation
Compare the following points with the official documentation.
Amazon Bedrock公式推論パラメータとAIF-C01 Domain 2.2を確認する。
Official references
What to confirm
サンプリング設定の効果と、実測評価が必要な理由を説明できる。
Key points
- Top-p
- Temperature
- Measurement
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
Top-p
A sampling setting that limits candidates to a smallest set reaching a cumulative probability.
Temperature
A sampling setting that changes the spread of probabilities used to choose candidate tokens.