As far as I know vector representations of word embeddings do not account for ambiguity. A single word can have different meanings f.e. "hot" can mean very good looking or very warm. Another example is "jaguar", that might refer to a car or an animal but the word is still represented by one vector while it has two or more meanings.
What methods take ambiguity into account? What approaches have been developed? What are the state of the art practices? A clear overview on this research topic would be appreciated. Thank you.