AIF-C01 Question 124
Single answerYou are designing a machine learning application that uses a large language model (LLM) to summarize customer feedback. The summaries generated often miss key details or include irrelevant information. Which approach in prompt engineering should you prioritize to improve the quality of the summaries?
- A
Provide clear and specific instructions within the prompt.
- B
Use a negative prompt to explicitly exclude irrelevant details.
- C
Rely on the model's latent space to infer the correct context automatically.
- D
Avoid adding context to the prompt to minimize its complexity.
Show answer and explanation
Correct answer: A
Explanation
The quality of outputs from large language models heavily depends on how prompts are crafted. In this scenario, providing clear and specific instructions in the prompt is the most effective way to guide the model to include key details and avoid irrelevant information. This approach directly addresses the issue and aligns with best practices in prompt engineering.
- A. Correct.
Providing clear and specific instructions in the prompt ensures the model understands the task and generates more accurate and relevant responses. This is a fundamental principle of prompt engineering.
- B. Incorrect.
While negative prompts can help exclude undesired outputs, they are less effective for ensuring that key details are included. This is not the best approach for this scenario.
- C. Incorrect.
The model's latent space is inherent to how the model processes information, but relying on it without proper prompt design will not address the issue. Prompt engineering explicitly guides the model's behavior.
- D. Incorrect.
Avoiding context in the prompt usually leads to vague or irrelevant outputs, as the model lacks sufficient guidance on what to focus on. Adding context is often essential for improving response quality.