1 paper
Yue Yang, Ajay Patel, Matt Deitke +8
Reasoning about images with rich text, such as charts and documents, is a critical application of vision-language models (VLMs). However, VLMs often struggle in these domains due t…