Your question is Generative Modeling with Imbalanced Classes. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
You are training a generative model on a labeled dataset where some classes are much rarer than others. You want the model to generate useful samples for all classes, not just the majority ones.
Given a dataset with imbalanced classes, how would you approach training a generative model?