Text this: An interpretable deep unfolding framework for multi-view representation learning.