Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper reframes long-horizon failure in world models by distinguishing between kinematic and dynamic errors, proposing a new diagnostic called imagined Kinematic-Consistency Error (iKCE). The authors apply this diagnostic to a DreamerV3 model trained on the DMC walker-walk task, revealing that the model's iKCE is significantly higher than that of real-physics rollouts, indicating a failure to accurately model dynamics. Notably, the model's iKCE remains constant even as the trained policy's performance collapses, highlighting a critical flaw in how these models imagine future states over extended horizons.
World models may be fundamentally misfiring by imagining future states kinematically, leading to significant performance drops without corresponding diagnostic signals.
Long-horizon failure in world models is conventionally attributed to compounding error, a generic framing that does not distinguish what kind of error compounds. We propose a kinematic-vs-dynamic reframing: world models tend to imagine kinematically rather than dynamically. We operationalize this as the imagined Kinematic-Consistency Error, a per-step diagnostic that measures how far a rollout departs from a closed-form kinematic null, paired with a perturbation protocol that tests whether iKCE responds when physical conditions cross a regime boundary. We instantiate the diagnostic on a released DreamerV3 checkpoint trained on DMC walker-walk, where imagined iKCE runs roughly two orders of magnitude above that of matched real-physics rollouts. Across a friction sweep that crosses the gait-collapse boundary, the model's iKCE stays statistically flat even as the trained policy's reward collapses through the same range, providing the kinematic-not-dynamic signature. The diagnostic distinguishes kinematic from dynamic imagination at horizons longer than the embodiment's gait period.