Evaluating Prompt Demonstrations for Sentiment Analysis
A developer is creating a system to classify customer reviews as 'Positive', 'Negative', or 'Neutral'. They have two options for the examples they will include in the instructions given to a large language model.
Option A: Review: "I love this phone, it's the best!" Sentiment: Positive Review: "This is the worst product I have ever bought." Sentiment: Negative
Option B: Review: "The camera quality is outstanding and the battery lasts all day." Sentiment: Positive Review: "The screen is prone to scratches and the software is buggy." Sentiment: Negative Review: "The phone was delivered in the original packaging." Sentiment: Neutral
Evaluate which set of examples (Option A or Option B) is more likely to lead to better performance for the classification task. Justify your evaluation by explaining the principles of providing effective demonstrations to a model.
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
A developer is building a system to sort customer feedback into 'Positive', 'Negative', or 'Neutral' categories. They provide the following complete text block to a large language model to classify the final review:
Classify the sentiment of the following review.
Review: "The battery life is amazing!" Sentiment: Positive
Review: "The screen scratches too easily." Sentiment: Negative
Review: "The product arrived on time." Sentiment:
Analyze this approach and identify the most significant issue that will likely lead to an unreliable classification for the final review.
Prompt Design for Sentiment Classification
Evaluating Prompt Demonstrations for Sentiment Analysis