Embeddings Quiz

5 questions Pass: 70% +25 pts (50 with Pro)

Quiz covering Transformers and Attention

What you'll be asked

  1. In the context of NLP, what is the primary purpose of an embedding vector?
  2. Why is 'cosine similarity' commonly used when comparing two word embeddings?
  3. What happens if two words appear in similar contexts during the training of an embedding model like Word2Vec?
  4. In Transformer-based models, what is the purpose of 'Positional Encodings' added to token embeddings?
  5. If you are using a pre-trained embedding model and notice that the model fails to represent the nuances of domain-specific jargon (e.g., medical or legal terms), what is the most technically sound way to resolve this?

Answers and explanations come after you submit.