Question

A marketing team uses a foundation model on Amazon Bedrock to classify customer feedback into categories like 'billing', 'shipping', and 'product quality'. With a plain instruction, the model's category names are inconsistent and sometimes invented. The team wants to improve accuracy without fine-tuning or retraining the model. What prompt engineering technique should they apply?

Accepted Answer

Include several example feedback messages paired with their correct category labels directly in the prompt (few-shot prompting)

Answer

Increase the temperature parameter so the model explores more category options

Answer

Reduce the maximum token limit so the model produces shorter, more focused responses

Answer

Switch to a larger context window model so more historical data can be stored