Generative AI Leader Practice Question: Techniques to Improve Generative AI Model Output
A media company uses a generative AI model to draft weekly newsletter articles from bullet points. The drafts are factually correct but read as terse and disjointed. Editors want smoother narrative flow without changing the underlying facts. Which technique should they apply first to improve the output?
⚠ Common exam trap
The trap here is assuming that any output-quality problem requires model fine-tuning, when prompt-level few-shot examples can fix a purely stylistic gap faster and at far lower cost.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Provide a few-shot prompt with two or three exemplar newsletter paragraphs demonstrating the desired narrative style.
The drafts are factually sound but stylistically weak, so the fastest effective fix is showing the model what good output looks like. Few-shot examples in the prompt communicate structure, tone, and transitions directly. Sampling changes and output-length limits affect randomness or size, not narrative quality, and fine-tuning is unnecessarily heavy for a style-only gap.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Fine-tune the model on the company's entire archive of past newsletters.
Why it's wrong here
Fine-tuning on an archive is heavier than necessary and risks overfitting to outdated topics or brand voice. It also requires substantial labeled data, compute, and time before any improvement appears. Since the issue is a stylistic prompt gap that few-shot examples can close immediately, full fine-tuning is disproportionate and slower to deliver the needed narrative flow.
- ✗
Increase the model's temperature setting so the model chooses less probable words.
Why it's wrong here
Raising temperature makes token selection more random, which can add variety but also increases factual drift and incoherence. This scenario already has factually correct content; the problem is structural flow, not word variety. Higher temperature would likely worsen disjointedness and introduce new errors, so it does not address the narrative cohesion the editors need.
- ✗
Reduce the maximum output tokens so the model must compress each article.
Why it's wrong here
Lowering the output token limit truncates generation and forces compression, which typically increases terseness rather than improving flow. The complaint is that drafts feel disjointed, not that they are too long. A shorter cap would cut transitions and connective phrasing first, making the narrative problem worse and potentially dropping required content.
- ✓
Provide a few-shot prompt with two or three exemplar newsletter paragraphs demonstrating the desired narrative style.
Why this is correct
Few-shot prompting supplies concrete examples that show the model the target structure, tone, and transitions. Because the facts are already correct, the examples teach style rather than content, which is exactly the gap. This is a low-cost, immediate prompt-level change that reliably improves flow without retraining or altering the factual bullet points.
Go deeper
Related to this question
About these practice questions
One of 1,008 original Generative AI Leader practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Google Cloud exam blueprint
This Generative AI Leader practice question is part of Courseiva's free Google Cloud certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the Generative AI Leader exam.