AWS Certified AI Practitioner

Explanation

When using generative AI models like those in Amazon Bedrock, the temperature parameter controls the randomness and creativity of the outputs:

• Lower temperature values (e.g., 0.2): Make the model more deterministic and focused on the most likely next tokens, resulting in more predictable and repetitive outputs.
• Higher temperature values (e.g., 0.9): Introduce more randomness and creativity, allowing the model to explore less likely but potentially more original and diverse outputs.

In this scenario, the marketing agency is experiencing repetitive and unoriginal outputs, which indicates the model is being too deterministic. Increasing the temperature to around 0.9 would encourage more creative variation in the generated brand-tagline ideas.

Why the other options are incorrect:

• Option A (Lower temperature to 0.2): This would make the outputs even more repetitive and deterministic, worsening the problem.
• Option C (Reduce top-p to 0.3): Top-p (nucleus sampling) controls the diversity of token selection. Reducing it would also make outputs more deterministic, not more creative.
• Option D (Add stop sequences): Stop sequences control when generation stops, not the creativity or originality of the content.

Key takeaway: Temperature is the primary parameter for controlling creativity vs. determinism in generative AI models. Higher temperatures promote creative variation, while lower temperatures produce more focused, predictable outputs.