2.1Standard
Generate an image from text.
Explain a noise-to-image process.
Choose two correct statements.
× Incorrect
Diffusion models learn adding and removing noise and can condition generation on text or image features; realism and factuality must be evaluated separately.
Detailed explanation
IncorrectDenoising always returns the same image in one deterministic step
Randomness and settings can change output.
CorrectA diffusion model learns a noising process and a denoising generation process
It can generate from a random state step by step.
IncorrectA diffusion output automatically verifies real-world facts
Realism and factuality are different.
CorrectText or image conditions can guide generated content
The result can be conditioned on input.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01 Domain 2.1の生成モデル、拡散モデル、マルチモーダルを確認する。
Official references
What to confirm
拡散モデルの生成過程と、生成結果の限界を説明できる。
Key points
- Noise
- Denoising
- Conditioning
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
Diffusion model
A generative model trained to reverse a process that gradually adds noise to data.
Denoising
Removing noise step by step to produce a sample from a learned distribution.
Question IDAWS-AIF-238
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