In vector semantics, the meaning of a word can change based on context. If the word "bark" is represented in two different contexts—one with "tree" and another with "dog"—what does this imply about the vector representation?

The distributional hypothesis suggests that words occurring in similar contexts tend to have similar meanings. If a new word "glorp" appears frequently with words like "delicious," "sauce," and "spicy," what can be inferred about "glorp"?

In a word embedding model, if the vector for "king" is represented as ( V_{king} ) and the vector for "queen" is ( V_{queen} ), which of the following operations would most likely yield a vector close to "woman"?

If two words, "car" and "automobile," have high cosine similarity in their vector representations, what does this imply about their meanings?

Which of the following forms of gradient descent updates the weight after computing the gradient over the entire dataset ?

You are working on finding the maximum likelihood estimate of seeing a bigram {“A”, “B”} where “B” and “A” are tokens and “B” follows