Emotion Prompting
Leverages emotional language to improve LLM accuracy and response quality by tapping into emotion-rich training data.
To use emotion prompting, you can add any of these phrases to your original prompt:
Role Prompting
Assigns roles to the LLM, creating more context-specific relevant responses for various tasks. By including in your prompt for the LLM a statement for them to take on the persona or role of a particular occupation or even person, you can alter the output in a way that may be more useful to your needs for the output.
Example:
You are a historian. Explain the significance of the industrial revolution. |
You are a salesman. Write a quick outreach email to [person] about partnering up. |
Rephrase and Response (RaR)
Asks the model to rephrase the prompt before answering, reducing ambiguity and improving clarity.
Apart from adding "Rephrase and expand the question, and respond" to the prompt, you can use many other variations for asking a model to rephrase the question:
All models can benefit from rephrasing questions, with more advanced models expected to gain a larger improvement.