Explaining Un-Taught Abilities
A large language model is trained on a massive dataset with a single objective: to predict the next word in a sequence. Despite this simple training goal, the model demonstrates a strong ability to paraphrase sentences while preserving their original meaning. In your own words, explain the underlying principle that accounts for this sophisticated, un-taught capability.
0
1
Tags
Ch.1 Pre-training - 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 research team develops a large language model trained exclusively on a single, simple task: predicting the next word in a sentence, using a vast corpus of text from the internet. During evaluation, they discover the model is highly effective at identifying grammatically incorrect sentences, a task it was never explicitly trained to perform. Which of the following statements provides the best analysis of this outcome?
A research team wants to build a model that is highly proficient at identifying the grammatical structure of sentences. Based on the principles of how large models acquire linguistic skills, their most effective primary strategy would be to train the model from scratch exclusively on a dataset of sentences that have been manually annotated with their grammatical structures.
Explaining Un-Taught Abilities