Other Prompt Types
Introduction This guide offers instructions for crafting prompts by incorporating different elements to improve functionality.
- Role Playing Prompt This type leverages the Language Model to create dynamic, engaging, and personalized interactions. It’s designed to simulate human-like conversations, guided by the unique personality traits of the Digital Human.
- Chatbot-Style Flow This prompt follows a structured, predefined conversational path, ensuring consistent and guided interactions. It is tailored for specific use cases requiring clear and predictable user journeys (coming soon).
- Dynamic and Adaptive: Conversations are flexible and can evolve based on the user's input and the context of the interaction.
- Human-Like Engagement: Designed to simulate natural, human-like conversations by incorporating traits like empathy, humor, and understanding.
- Personalized Experience: Interactions feel unique as the responses are influenced by the "personality" of the digital assistant, which can adapt to the user's mood or preferences.
- Unscripted: There is no rigid conversational flow, allowing for open-ended and creative exchanges.
- Use Cases: Ideal for customer support, personal assistants, or scenarios where human touch and flexibility are valued.
- Structured and Predefined: Conversations follow a set sequence or flowchart, ensuring specific paths and outcomes are maintained.
- Consistency: Responses are predictable and guided, making interactions reliable for standardized processes.
- Task-Oriented: Designed for specific use cases where the objective is to achieve a clear goal, such as booking a ticket, answering FAQs, or collecting data.
- Limited Personalization: While responses may address user input, they are typically not as adaptive or influenced by personality traits.
- Use Cases: Best for transactional tasks, customer service workflows, or scenarios requiring compliance and consistency, such as banking chatbots or e-commerce assistants.
Aspect | Personality-Based Conversation | Chatbot-Style Flow |
|---|---|---|
Flexibility | Dynamic and adaptive | Structured and predefined |
Engagement Style | Human-like, personalized | Guided, task-focused |
Complexity | Handles open-ended queries | Handles specific, pre-set tasks |
Use Cases | Creative, exploratory, emotional | Transactional, procedural |
Response Style | Unique, varied | Consistent, predictable |
Role-Playing Prompt
Personality prompt:
"You are an expert eLearning teacher who excels in engaging virtual classrooms and designing effective online learning experiences. Your students come from diverse backgrounds, and your goal is to ensure maximum understanding and retention of the subject matter. Please create an interactive lesson plan for a one-hour online session on [Insert Topic].
Specific instructions:
The lesson plan should include:
- An engaging introduction that hooks the students.
- Three key learning objectives for the session.
- Activities or multimedia elements to make the session interactive (e.g., polls, discussions, videos).
- A short quiz or assessment at the end to reinforce learning.
- Suggestions for follow-up resources for students to explore after the session.
Ensure the lesson is suitable for [Insert target audience details, e.g., high school students, working professionals], and prioritize accessibility and inclusivity in your approach."
Example: Ask the following questions individually, wait for the user's response, acknowledge their answer briefly, and then proceed to the next question until the survey is complete.
Personality prompt
Act as an interactive chatbot designed to conduct a 5-question survey. Your task is to ask one question at a time, wait for the user's response, acknowledge their answer briefly, and then proceed to the next question. After all 5 questions are answered, provide a thoughtful and comprehensive final response based on the user’s answers.
General Guidelines:
Assistant: I’m going to ask you 5 questions one by one. After each question, please provide your answer. Once all 5 questions are answered, I will give you a final response based on your answers.
Question 1: [Ask your first question here] (Wait for user response.)
Question 2: [Ask your second question here] (Wait for user response.)
Question 3: [Ask your third question here] (Wait for user response.)
Question 4: [Ask your fourth question here] (Wait for user response.)
Question 5: [Ask your fifth question here] (Wait for user response.)
Assistant (Final Response): [Provide a comprehensive response based on the user’s answers to all 5 questions.]
