Prompt Template for Answer Generation from Context
A prompt can be formatted to instruct a Large Language Model to use retrieved reference information while generating an answer in its own words. An example format is:
You are an expert that can generate answers to input queries. You have now been provided with a query and the corresponding context information. Please generate an answer based on this context information. Note that you need to provide the answer in your own words, not just copy from the context provided. Context information: {IR-result} Query: {query}
0
1
Tags
Foundations of Large Language Models
Ch.3 Prompting - Foundations of Large Language Models
Foundations of Large Language Models Course
Computing Sciences
Related
Factual Accuracy for a Corporate Chatbot
A company implements a chatbot to answer employee questions based on a private, internal knowledge base of company policies. An employee asks the chatbot, 'What is the company's policy on parental leave?' The chatbot provides a detailed, accurate, and up-to-date answer. The employee then asks, 'What are the best public parks near the main office?' The chatbot responds, 'Based on the provided documents, I cannot find any information about public parks.' Which statement best analyzes the chatbot's behavior in the second interaction?
A system is designed to answer a user's question using a large, private collection of documents. The system first finds relevant information within the documents and then uses that information to construct a natural language answer. Arrange the following steps to reflect the correct operational sequence of this system.
Prompt Template for Answer Generation from Context