Multiple Choice

A development team has a small, high-quality dataset for training a sentiment analysis model. To improve the model's performance without collecting more user data, they use a powerful, general-purpose language model to paraphrase each existing example, generating five new variations for every original sentence while preserving the sentiment label. This process of creating synthetic training examples is most directly analogous to which traditional machine learning practice, and why?

0

1

Updated 2025-10-05

Contributors are:

Who are from:

Tags

Ch.4 Alignment - 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