Recent video diffusion models have achieved impressive capabilities as large-scale generative world models. However, these models often struggle with fine-grained physical consistency, exhibiting physically implausible dynamics over time. In this work, we present \textbf{Phys4D}, a pipeline for learning physics-consistent 4D world representations from video diffusion models. Phys4D adopts \textbf{a three-stage training paradigm} that progressively lifts appearance-driven video diffusion models into physics-consistent 4D world representations. We first bootstrap robust geometry and motion representations through large-scale pseudo-supervised pretraining, establishing a foundation for 4D scene modeling. We then perform physics-grounded supervised fine-tuning using simulation-generated data, enforcing temporally consistent 4D dynamics. Finally, we apply simulation-grounded reinforcement learning to correct residual physical violations that are difficult to capture through explicit supervision. To evaluate fine-grained physical consistency beyond appearance-based metrics, we introduce a set of \textbf{4D world consistency evaluation} that probe geometric coherence, motion stability, and long-horizon physical plausibility. Experimental results demonstrate that Phys4D substantially improves fine-grained spatiotemporal and physical consistency compared to appearance-driven baselines, while maintaining strong generative performance. Our project page is available at https://sensational-brioche-7657e7.netlify.app/

翻译：近年来，视频扩散模型作为大规模生成式世界模型展现出令人印象深刻的能力。然而，这些模型往往难以实现细粒度的物理一致性，在时间维度上呈现出违反物理规律的动态。本文提出\textbf{Phys4D}，一种从视频扩散模型中学习物理一致性4D世界表征的流水线。Phys4D采用\textbf{三阶段训练范式}，逐步将外观驱动的视频扩散模型转化为物理一致的4D世界表征。首先，通过大规模伪监督预训练引导出稳健的几何与运动表征，为4D场景建模奠定基础；然后，利用模拟生成的数据进行基于物理的有监督微调，强制实现时间一致性4D动态；最后，应用基于模拟的强化学习来修正显式监督难以捕获的残留物理违规。为评估超越外观指标的细粒度物理一致性，我们引入一套\textbf{4D世界一致性评估}体系，用于探测几何连贯性、运动稳定性及长期物理合理性。实验结果表明，与外观驱动基线方法相比，Phys4D在显著提升细粒度时空与物理一致性的同时，保持了强大的生成性能。项目页面见https://sensational-brioche-7657e7.netlify.app/