RISE：世界行動模型的自適應想像力

摘要

世界行動模型（WAMs）透過將未來世界演化納入動作生成來改善規劃，然而現有方法對每個場景都分配固定的想像預算。我們提出 RISE（透過選擇性展開精煉想像），這是一個系統層級的自適應想像框架，依據繼續展開的預期規劃效益來做出序列化的 Roll/Stop（展開/停止）決策。在每一步中，潛在評估器（Latent Evaluator）估計當前前綴所揭示的風險，以及若繼續想像能為規劃帶來多少改善；同時，展開門控（Rollout Gate）將此預期效益與額外的計算成本進行權衡。由於事實駕駛日誌僅揭露單一已實現的未來，我們進一步建構了 CounterDrive，一個具有多樣化結果與風險等級的反事實資料集，以豐富未來動態並提供局部化風險監督。每個保留的樣本都經過專家驗證，並對軌跡有效性、事件發生點與因果類別進行標註，為安全關鍵的世界建模研究提供了可重複使用的資源。在 NAVSIM 與 nuScenes 上的實驗顯示，RISE 在減少不必要展開的同時達到了最佳的整體規劃效能，額外的遷移結果也支持其跨 WAM 架構的即插即用通用性。

English

World Action Models (WAMs) improve planning by incorporating future world evolution into action generation, yet existing methods allocate a fixed imagination budget to every scene. We propose RISE (Refining Imagination through SElective Rollout), a system-level adaptive imagination framework that makes sequential Roll/Stop decisions according to the expected planning benefit of continued rollout. At each step, a Latent Evaluator estimates the risk revealed by the current prefix and how much planning could improve if imagination continues, while a Rollout Gate weighs this expected benefit against additional computation cost. Since factual driving logs expose only one realized future, we further construct CounterDrive, a counterfactual dataset with diverse outcomes and risk levels, to enrich future dynamics and provide localized risk supervision. Each retained sample undergoes expert verification and annotation of trajectory validity, incident onset, and causal category, providing a reusable resource for safety-critical world-modeling research. Experiments on NAVSIM and nuScenes show that RISE achieves the best overall planning performance while reducing unnecessary rollout, with additional transfer results supporting its plug-in generality across WAM architectures.