Probing Foundation Models for World Models

This paper investigates whether foundation models truly acquire a deeper understanding of underlying "world models" beyond mere accurate sequence prediction. Researchers introduce an "inductive bias probe" to evaluate how these models adapt to new tasks based on postulated world models, such as Newtonian mechanics for orbital trajectories or game rules for Othello. The findings suggest that while foundation models excel at their primary training objectives, they often fail to develop strong inductive biases toward the actual governing principles. Instead, they appear to rely on task-specific heuristics or coarsened state representations, leading to a lack of generalizability.