Evaluating Data Strategies for Model Fine-Tuning
A small startup is building a language model to act as a highly specialized legal assistant for contract review. They have a limited budget and a tight deadline. They are considering two different approaches for creating their fine-tuning dataset. Based on the principles of effective model fine-tuning, which approach would you recommend and why?
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
Evaluation in Bloom's Taxonomy
Cognitive Psychology
Psychology
Social Science
Empirical Science
Science
Related
Evaluating Data Strategies for Model Fine-Tuning
A development team is fine-tuning a language model for a specialized medical question-answering task where accuracy is critical. They have two potential datasets: Dataset A consists of 100,000 unfiltered Q&A pairs scraped from various online health forums. Dataset B consists of 5,000 Q&A pairs carefully curated and verified for accuracy by medical experts. Which statement best evaluates the most effective approach for the team?
Optimizing Fine-Tuning Data Strategy