OneWorld：在世界模型中学习跨动作的物理一致性
OneWorld: Learning Consistent Physics Across Actions in World Models
查看机构详情
- Wuhan University（武汉大学）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
针对视频世界模型跨动作预测物理不一致的问题，提出OneWorld共享机制反事实生成框架，通过物理机制解释与共享证据约束，提升跨干预物理一致性并保持单轨迹预测质量。
中文摘要 AI 辅助
动作条件下的视频世界模型旨在预测不同动作下的场景演化，这一能力对于动态环境中的可靠规划、决策和交互至关重要。然而，从同一初始场景独立生成的未来可能各自看似合理，却暗示着不兼容的物理属性，如摩擦或质量。这种不一致可能导致跨干预的预测相互矛盾，使模型难以对底层世界保持连贯理解，并限制其在规划和决策中的可靠性。为解决这些问题，我们提出OneWorld，一种共享机制的反事实生成框架，该框架在共同潜在物理机制下联合建模多个动作条件下的未来。物理机制解释器首先从每个动作-结果分支推断潜在机制的分布。随后，这些分布被聚合为共享世界证据，该证据捕获各分支是否承认共同的物理解释，同时考虑信息较少分支中的不确定性。该证据约束流训练并指导采样，在保持不同动作所引发的不同结果的同时，促进底层物理机制的一致性。我们进一步在受控环境中引入多干预评估协议，遵循ACWM-Phys的交互设置，以评估生成的未来是否可由相同物理参数联合解释，同时结合标准单轨迹预测质量度量。在这些环境中的实验表明，OneWorld在保持有竞争力的单轨迹预测质量的同时，提高了跨干预的物理一致性。
英文摘要
Action-conditioned video world models aim to predict scene evolution under different actions, a capability that is essential for reliable planning, decision-making, and interaction in dynamic environments. However, futures generated independently from the same initial scene may each appear plausible while implying incompatible physical properties, such as friction or mass. This inconsistency can lead to contradictory predictions across interventions, making it difficult for the model to maintain a coherent understanding of the underlying world and limiting its reliability for planning and decision-making. To address these issues, we propose OneWorld, a shared-mechanism counterfactual generation framework that jointly models multiple action-conditioned futures under a common latent physical mechanism. A physical mechanism interpreter first infers a distribution over latent mechanisms from each action-outcome branch. These distributions are then aggregated into shared-world evidence, which captures whether the branches admit a common physical explanation while accounting for uncertainty in less informative branches. This evidence constrains flow training and guides sampling, encouraging consistency in the underlying physical mechanism while preserving the distinct outcomes induced by different actions. We further introduce a multi-intervention evaluation protocol in controlled environments, following the interaction settings of ACWM-Phys, to assess whether generated futures can be jointly explained by the same physical parameters, alongside standard measures of single-rollout prediction quality. Experiments in these environments show that OneWorld improves cross-intervention physical consistency while maintaining competitive single-rollout prediction quality.
