Answer: Add clearer instructions and define output format

**Explanation:** When using Amazon Bedrock or any generative AI model, vague outputs often result from insufficiently detailed prompts. To improve output quality: 1. **Clearer instructions**: Provide specific guidance on what information to extract, what format to use, and what level of detail is expected. 2. **Define output format**: Specify exactly how the financial summary should be structured (e.g., bullet points, tables, specific sections). 3. **Context and examples**: Including examples of desired outputs can significantly improve model performance. **Why other options are incorrect:** - **A) Use a shorter prompt**: Shorter prompts typically provide less context and guidance, which would likely make outputs even more vague. - **C) Randomize wording for creativity**: Randomization introduces inconsistency and unpredictability, which is counterproductive for structured financial summaries. - **D) Reduce temperature to zero**: While lowering temperature can make outputs more deterministic, it doesn't address the core issue of vague instructions. Temperature controls randomness, not clarity of instructions. **Best Practice**: When working with Amazon Bedrock, use prompt engineering techniques like: - Providing clear role definitions ("You are a financial analyst...") - Specifying exact output formats - Including examples of good vs. bad outputs - Using system prompts to set context and constraints

Author: Ritesh Yadav