Question: When using contextual embeddings from a model like GPT , how would the word right in Turn right at the corner differ from He
When using contextual embeddings from a model like GPT how would the word "right" in "Turn right at the corner" differ from He has the right to vote"?
Group of answer choices
The embedding remains the same since "right" is the same word.
The embedding is slightly altered due to sentence length.
The embedding is based on the frequency of "right" in the training corpus.
The embedding changes based on the different contexts of usage.
Step by Step Solution
There are 3 Steps involved in it
1 Expert Approved Answer
Step: 1 Unlock
Question Has Been Solved by an Expert!
Get step-by-step solutions from verified subject matter experts
Step: 2 Unlock
Step: 3 Unlock
