Analyzing Trade-offs in In-Context Learning Strategies
A developer is building an application that uses a large language model to perform a novel task. They are considering using zero-shot, one-shot, or few-shot prompting. Analyze the primary trade-offs between these three approaches in terms of task performance, prompt complexity, and computational cost. In your analysis, explain how the number of examples provided in the prompt influences these factors.
0
1
Tags
Ch.2 Generative Models - Foundations of Large Language Models
Foundations of Large Language Models
Foundations of Large Language Models Course
Computing Sciences
Analysis in Bloom's Taxonomy
Cognitive Psychology
Psychology
Social Science
Empirical Science
Science
Related
A user provides the following input to a large language model to get it to classify the sentiment of movie reviews. The model has not been specifically trained or fine-tuned on this particular task.
Classify the sentiment of the movie review. Review: "The plot was predictable and the acting was wooden." Sentiment: Negative Review: "An absolute masterpiece of modern cinema!" Sentiment:Which in-context learning approach does this prompt demonstrate?
Choosing an In-Context Learning Strategy
Analyzing Trade-offs in In-Context Learning Strategies