Question 27
Domain 5: Describe knowledge mining and generative AI workloads on AzureWhat is the purpose of vector-based embeddings?
Correct answer: A
Explanation
Vector-based embeddings map text tokens into numeric vectors so similar meanings are placed near each other in vector space. This lets models capture "semantic meaning of text tokens" rather than just their surface form, which is why they are used for meaning-based comparison and retrieval.
Why each option is right or wrong
A. To represent semantic meaning of text tokens.
Vector embeddings are used to convert discrete tokens into dense numerical vectors so the model can work with meaning rather than only token identity. In practice, tokens with related contexts are placed closer together in the embedding space, which is what allows downstream models to compare and process semantic relationships between words, phrases, and documents.
B. To create tokens that include multiple representations of a word in different languages.
C. To correct misspellings in the training data.