Researchers from the University of Oxford and the National University of Singapore have introduced a framework called Mental World Modeling (MWM) in their study, "Mental World Modeling." The framework integrates mental variables directly into the world state, rather than treating them as post-hoc explanations. The research addresses a key issue: world models that only track objects, locations, and motion may produce predictions that are physically plausible but behaviorally incorrect. For example, if Person A puts a cup into a cabinet and Person B does not see it, a model that only knows physical facts would incorrectly predict that B will look in the cabinet, while a model that understands the human mind would correctly predict that B believes the cup is still on the table. The researchers argue that most existing world models focus primarily on physical aspects, treating humans as objects that move and perform actions, without incorporating their internal states. However, human behavioral decisions emerge from the interaction between external environments and internal psychological-social factors. The study notes that scenarios such as service robots determining user confusion or medical assistants considering patient cognition require intelligent systems to track agents' mental and social states. As of press time, MWM has ranked #1 on Hugging Face's Daily Papers.