Consider the following instruction given to a language model: 'I read an article about the economic impact of renewable energy. Make it shorter.' Analyze this instruction and identify two specific weaknesses that make it ineffective for a text summarization task. Then, provide one example of a well-structured, improved instruction that addresses these weaknesses.

Google

A text summarization task is typically initiated with a prompt that instructs a language model. This prompt usually contains a direct command, like 'Summarize the following article,' followed by the source text that needs to be condensed. For example: 'Article: In recent years, solar energy has seen unprecedented growth...'

Example of a Text Summarization Prompt

A user wants to get a concise, one-paragraph overview of a long article using an AI model. Which of the following inputs is structured most effectively to instruct the model to perform this specific task?

Critique and Refine a Summarization Prompt

Analyze the following scenario and explain why the user's instruction failed to produce a proper summary of the entire article.

Learn Before

Related