A development team trains a large language model on a vast, unprocessed collection of text from the internet. During testing, they find the model frequently produces outputs that are nonsensical, contain harmful stereotypes, and include fabricated information. Which of the following strategies should the team prioritize to most effectively address the root cause of these issues before their next training attempt?
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 development team trains a large language model on a vast, unprocessed collection of text from the internet. During testing, they find the model frequently produces outputs that are nonsensical, contain harmful stereotypes, and include fabricated information. Which of the following strategies should the team prioritize to most effectively address the root cause of these issues before their next training attempt?
Evaluating a Data Cleaning Strategy for LLM Training
A machine learning team is preparing a large text dataset to train a new language model. Arrange the following data processing steps into a logical and effective sequence.