We've introduced a new abstraction and a document Retriever called ContextualCompressionRetriever, which facilitates the post-processing of retrieved documents by compressing them based on the context of the given query, making it easy to extract only relevant information from the retrieved documents.