Article

AnyGoal: Evaluating World Models' Compositional Generalization Beyond Expert Trajectories

The study introduces the AnyGoal benchmark and open-source library to evaluate world models' compositional generalization on goal-shifted task variants. The task design ensures that individual state factors and actions remain within the training support, but their combinations are underrepresented in expert trajectories. Experiments across multiple environments and world model families show that current models consistently fail under these recombination scenarios, and the authors further analyze where failures occur at the dynamics and representation levels.

Part of the event: AnyGoal benchmark released; compositional generalization in world models generally fails

This page provides a summary and source link only.