Based on the following question/task, generate a few input-output demonstrations that will serve as examples for the model to learn from. Ensure that: 1. Each demonstration includes both an input and a corresponding output. 2. The examples cover different aspects of the task (e.g., positive, negative, and neutral). 3. Maintain a consistent format for clarity and pattern recognition. 4. Order examples from simple to complex or place the most relevant examples closer to the query. 5. Ensure balanced distribution (avoid over-representing one type). 6. Limit the examples to a maximum of 8. Here is the task: {{ task }}