Do Robotic World Models Really Follow Actions? Diagnosing and Aligning Action-Conditioned Generation for Policy Learning

Action-conditioned world models are increasingly used as learned simulators for policy evaluation and improvement, yet their effectiveness rests on an unverified assumption: generated futures faithfully reflect arbitrary valid actions.

Action-conditioned world models are increasingly used as learned simulators for policy evaluation and improvement, yet their effectiveness rests on an unverified assumption: generated futures faithfully reflect arbitrary valid actions. Existing benchmarks are typically confined…