The top half of the diagram is based on the Langlands program, which predicts that any motivic object corresponds to an automorphic object via their L-functions.