Q28 — AWS AIF-C01 Ch.2

Question 28 of 100 | ← Chapter 2

A company is using a pre-trained large language model (LLM) to build a chatbot for product recommendations. The company wants the LLM’s output to be concise and follow a specific language style. Which solution ensures the LLM’s response quality meets the company’s expectations?

• A. Adjust the prompt. ✓
• B. Select a different-sized LLM.
• C. Increase temperature.
• D. Increase Top-K value.

Correct Answer: A. Adjust the prompt.

Explanation

To ensure a pre-trained LLM generates concise, stylistically consistent output, prompt engineering—i.e., carefully crafting the prompt—is the most effective method. It explicitly guides the model on length, tone, and formatting. Selecting a different model size affects capacity and knowledge but not stylistic control. Increasing temperature increases randomness and reduces determinism, conflicting with consistency goals. Increasing Top-K broadens token sampling without guaranteeing conciseness or style adherence. Thus, prompt adjustment is optimal.