Learn Before
A company is developing an automated system to classify thousands of customer support emails per day into one of three categories: 'Urgent', 'Standard', or 'Spam'. They are considering two prompting strategies for their large language model.
- Strategy X: For each email, the system sends a long prompt that includes a detailed, 200-word definition of each category, several examples, and the full text of the customer email.
- Strategy Y: For each email, the system sends a short prompt containing only a brief instruction ('Classify this email:') and the full text of the customer email.
From a computational cost perspective, which strategy is the most suitable for this high-volume, repetitive task, and why?
0
1
Tags
Ch.3 Prompting - Foundations of Large Language Models
Foundations of Large Language Models
Foundations of Large Language Models Course
Computing Sciences
Evaluation in Bloom's Taxonomy
Cognitive Psychology
Psychology
Social Science
Empirical Science
Science
Related
LLM-Powered Email Categorization System
A company is developing an automated system to classify thousands of customer support emails per day into one of three categories: 'Urgent', 'Standard', or 'Spam'. They are considering two prompting strategies for their large language model.
- Strategy X: For each email, the system sends a long prompt that includes a detailed, 200-word definition of each category, several examples, and the full text of the customer email.
- Strategy Y: For each email, the system sends a short prompt containing only a brief instruction ('Classify this email:') and the full text of the customer email.
From a computational cost perspective, which strategy is the most suitable for this high-volume, repetitive task, and why?
Chatbot Prompting Strategy Analysis