HiFOR: Hierarchical Foresight with Action-Guided World Grounding for Vision-Language-Action Model

HiFOR is a hierarchical foresight framework with action-guided world retrieval. Hierarchical Foresight Prediction jointly predicts short-term state evolution, semantic milestones, and final goal states through a Hierarchical Dependency Mask. Structure-Aware Latent World Modeling organizes future latent representations along the horizon, view, and spatial dimensions.