SG-WAM：几何感知策略空间中的自引导世界建模

摘要

世界动作模型（WAMs）将动作生成与未来状态预测相耦合。其有效性取决于未来动态能否在一个既与动作生成对齐、又具备足够几何感知能力以刻画动作在何处以及如何改变场景的空间中被建模。现有WAMs通常仅部分满足这一需求：要么依赖感知负担较重的观测空间目标，要么依赖未针对动作相关性与几何结构进行联合组织的辅助潜在空间。我们提出SG-WAM，一种自引导框架，直接在策略派生的表示空间中学习几何感知的动作条件动态。SG-WAM引入可学习的动态令牌（dynamics tokens）以及自引导世界预测器（Self-Guided World Predictor），后者在介入的机器人动作条件下预测令牌的未来潜在状态。预测目标由同一策略主干网络的指数移动平均副本生成，从而在动作专家所使用的表示族内提供稳定监督。几何监督进一步结构化了策略的图像令牌表示，为动态令牌提供空间锚定的上下文，并产生一个兼具动作相关性与几何感知能力的未来对齐空间。潜在未来预测、几何锚定与流匹配动作生成在统一框架中以端到端方式联合优化。SG-WAM基于0.9B参数模型，无需大规模具身预训练，在LIBERO上取得98.5%的平均成功率，在LIBERO-Plus上取得73%的平均成功率，并在分布内与分布外真实世界评估中均优于强基线。

English

World Action Models (WAMs) couple action generation with prediction of future states. Their effectiveness depends on whether future dynamics are modeled in a space that is both aligned with action generation and sufficiently geometry-aware to capture where and how actions change the scene. Existing WAMs typically satisfy only part of this requirement, relying on either perceptually heavy observation-space targets or auxiliary latent spaces that are not jointly structured for action relevance and geometry. We propose SG-WAM, a self-guided framework that learns geometry-aware action-conditioned dynamics directly in the policy-derived representation space. SG-WAM introduces learnable dynamics tokens and a Self-Guided World Predictor that forecasts their future latent states conditioned on intervening robot actions. Prediction targets are generated by an exponential moving average copy of the same policy backbone, providing stable supervision within the representation family used by the action expert. Geometric supervision further structures the policy image-token representations, providing spatially grounded context for the dynamics tokens and yielding a future-alignment space that is both action-relevant and geometry-aware. Latent future prediction, geometric grounding, and flow-matching action generation are jointly optimized end-to-end in a unified framework. Built on a 0.9B model without large-scale embodied pretraining, SG-WAM achieves 98.5% average success on LIBERO and 73% on LIBERO-Plus, while outperforming strong baselines in both in-distribution and out-of-distribution real-world evaluations.