Why prompt design determines output quality in Prompt Engineering / GenAI - Challenge Your Understanding

Challenge - 5 Problems

🧠 Conceptual

intermediate

How does prompt clarity affect AI output?

Which statement best explains why clear prompts lead to better AI-generated answers?

A Clear prompts make the AI run faster by using less memory.

B Clear prompts increase the AI's training data size automatically.

C Clear prompts allow the AI to ignore irrelevant parts of the input.

D Clear prompts reduce ambiguity, helping the AI understand exactly what is asked.

💡 Hint

Think about how the AI interprets your question.

Clear prompts reduce confusion and ambiguity, so the AI can generate more accurate and relevant responses.

❓ Predict Output

intermediate

Output difference from prompt specificity

Given the prompt "List three fruits." and the prompt "List three tropical fruits.", what difference in output would you expect from a language AI?

A The first prompt lists any fruits; the second lists only tropical fruits.

B Both prompts list the same fruits in the same order.

C The first prompt lists tropical fruits only; the second lists all fruits.

D The AI will produce an error for the second prompt.

💡 Hint

Consider how adding details changes the AI's focus.

Adding 'tropical' narrows the AI's response to fruits from tropical regions, making the output more specific.

❓ Model Choice

advanced

Choosing model size for prompt sensitivity

You want an AI model that responds well to detailed prompts and can generate nuanced answers. Which model type is best?

A A large model trained on diverse data.

B A small model with limited parameters.

C A model trained only on numeric data.

D A model with no training, just random outputs.

💡 Hint

Think about model capacity and training diversity.

Larger models trained on diverse data understand complex prompts better and produce richer outputs.

❓ Hyperparameter

advanced

Effect of temperature on prompt output

In generative AI, what does increasing the 'temperature' hyperparameter do to the output when given the same prompt?

A Freezes the output to always be the same.

B Causes the model to ignore the prompt.

C Makes the output more random and creative.

D Makes the output shorter and more precise.

💡 Hint

Temperature controls randomness in generation.

Higher temperature increases randomness, leading to more varied and creative outputs.

❓ Metrics

expert

Evaluating prompt effectiveness with BLEU score

You have two prompts generating text outputs. You want to measure which prompt produces text closer to a reference answer. Which metric is best to use?

A Mean squared error on token IDs.

B BLEU score measuring n-gram overlap.

C Counting the number of words in output.

D Measuring the time taken to generate output.

💡 Hint

Think about comparing text similarity.

BLEU score measures how closely generated text matches a reference by comparing word sequences.