Abstract
Motion recognition is one of the basic cognitive capabilities of many life forms, yet identifying motion of physical entities in natural language have not been extensively explored empirically. We present the first Motion in Text dataset, a human-annotated collection of text sentences describing physical occurrence of motion, with annotated physical entities in motion. We describe the annotation process of our dataset, analyze its scale and diversity, and report results of several baseline models. We also present the challenges of annotating motion in natural language and share publicly the dataset as a new challenge to the research community.