SKIP: Sparse Keyframe Interpolation Paradigm for Efficient Embodied World Models 事件

SKIP: Sparse Keyframe Interpolation Paradigm for Efficient Embodied World Models arXiv:2606.00664v1 Announce Type: cross Abstract: Embodied world models have emerged as a promising paradigm in robotics by predicting how robot actions affect the surrounding scene. However, the rollout inference remains computationally expensive in pixel space, as long-horizon manipulation videos typically have to be generated frame by frame. This cost cannot be easily reduced by indiscriminately dropping frames,