This line of research focuses on the intersection of computer vision and natural language understanding. In particular, we study tasks that require visual input in the form of images or video as well as linguistic input in the form of text or audio. We aim at designing novel methods and algorithms that can jointly process these diverse and dissimilar types of information.