Images are central to our understanding of and learning about our world. We argue that visual training needs to be improved in the higher education system to enhance the potential of visual thinking to mediate productively our relationship with the context we inhabit. Initiatives from the social sciences and humanities since the end of the twenty…

In real-world settings, vision language models (VLMs) should robustly handle naturalistic, noisy visual content as well as domain-specific language and concepts. For example, K-12 educators using digital learning platforms may need to examine and provide feedback across many images of students' math work. To assess the potential of VLMs to support…