Evaluating a Task Description for Entity Linking
An engineer is trying to instruct a language model to identify all phrases that refer to the same entity in a text. They are using a prompt structure that includes a placeholder for a specific instruction: Your task is to {task-description}.
The engineer uses the following instruction for the placeholder: find the pronouns.
Briefly explain why this instruction is insufficient for the overall goal and provide a more complete and effective instruction to replace it.
0
1
Tags
Ch.3 Prompting - Foundations of Large Language Models
Foundations of Large Language Models
Computing Sciences
Foundations of Large Language Models Course
Analysis in Bloom's Taxonomy
Cognitive Psychology
Psychology
Social Science
Empirical Science
Science
Related
An AI engineer is creating a prompt to instruct a language model. The goal is for the model to process a document and identify all the different phrases that refer to the same person or object. For example, in the sentence "Dr. Evelyn Reed published her findings; the scientist was praised for the work," the model should recognize that "Dr. Evelyn Reed," "her," and "the scientist" all refer to the same individual.
The engineer uses the following prompt structure:
You will be provided with a text. Your task is to {task-description}.Which of the following options for
{task-description}is the most precise and effective for achieving the engineer's goal?Debugging a Prompt for Entity Linking
Evaluating a Task Description for Entity Linking