快速LeWorldModel

摘要

聯合嵌入預測架構（JEPAs，Joint-Embedding Predictive Architectures），包括近期的LeWorldModel（LeWM），已成為一種有前景的無重建視覺世界模型基礎。然而，在視覺規劃中，LeWM透過反覆應用局部單步潛在轉換模型來評估候選動作序列。這種自迴歸滾動（autoregressive rollout）使規劃計算成本高昂，並且隨著預測視野（horizon）的增長，會使預測軌跡暴露於累積的潛在誤差中。我們提出Fast LeWorldModel（Fast-LeWM），這是一種快速潛在世界模型，它以動作前綴預測（action-prefix prediction）取代了重複的局部滾動。給定當前潛在狀態和一組候選動作序列，Fast-LeWM編碼其前綴，並並行預測執行這些前綴後所達到的未來潛在狀態。透過將動作前綴作為基本預測單元，Fast-LeWM直接建模了動作效應在不同視野範圍內不同程度的累積。這種前綴層級的監督迫使模型學習狀態如何在不同的動作前綴下持續演化，而不僅僅是擬合單步狀態轉換。在規劃過程中，預測器可以直接使用編碼動作序列中的最後一個前綴標記來評估對應的未來潛在狀態，而無需明確地滾動經過每個中間的想像狀態。在多個任務中，Fast-LeWM相較於LeWM在平均成功率上有所提升，同時大幅減少了規劃時間，並實現了更低的開環潛在損失（open-loop latent loss），且該損失隨著滾動視野的增加而增長速度顯著變慢。

English

Joint-Embedding Predictive Architectures (JEPAs), including recent LeWorldModel (LeWM), have become a promising foundation for reconstruction-free visual world models. For visual planning, however, LeWM evaluates candidate action sequences by repeatedly applying a local one-step latent transition model. This autoregressive rollout makes planning computationally expensive and exposes the predicted trajectory to accumulated latent errors as the horizon grows. We propose Fast LeWorldModel (Fast-LeWM), a fast latent world model that replaces repeated local rollout with action-prefix prediction. Given the current latent and a candidate action sequence, Fast-LeWM encodes its prefixes and predicts the future latents reached after executing those prefixes in parallel. By making action prefixes the basic prediction unit, Fast-LeWM directly models action effects accumulated to different extents over multiple horizons. This prefix-level supervision forces the model to learn how states continuously evolve under different action prefixes, rather than only fitting one-step state transitions. During planning, the predictor can use the last prefix token from the encoded action sequence to evaluate the corresponding future latent without explicitly rolling through each intermediate imagined state. Across multiple tasks, Fast-LeWM improves average success over LeWM while substantially reducing planning time, achieving lower open-loop latent loss whose growth becomes significantly slower as the rollout horizon increases.