Which of the following correctly pairs each model with its category — Generative or Discriminative?

Generative models learn the joint probability distribution P(X, Y) — effectively modelling how the data itself, including its class labels, was generated — and can be used to generate new synthetic samples. Naive Bayes is generative because it explicitly models P(X|Y) and P(Y) and combines them via Bayes' theorem. Discriminative models, by contrast, learn the conditional probability P(Y|X) directly, focused purely on distinguishing between classes without modelling how the input data arose. Logistic Regression is a classic discriminative model, since it directly models the decision boundary/class probability given the input features.