ENACT: Evaluating Embodied Cognition with World Modeling of Egocentric Interaction

This article introduces ENACT, a framework for evaluating embodied cognition. The focus is on using world modeling to understand egocentric interactions. The research likely explores how AI agents can learn and reason about the world from their own perspective, which is a key aspect of embodied intelligence. The use of 'ArXiv' as the source suggests this is a pre-print research paper, indicating it's likely a novel contribution to the field.