AI 中文总结
本研究针对仅用单摆视频训练的DreamerV3，通过无标签搜索发现其学习到类能量不变量，投影隐态回初始水平集可降低保守模型滚动误差，揭示世界模型存在从像素学物理约束却在想象时违反的失效模式。
AI 中文摘要
世界模型能够预测视频，却无需学习其可靠保留的动力学。我们测试了仅在单摆视频上训练的冻结DreamerV3，是否会学习到一个其自身隐态转移视为近似守恒的标量。无标签搜索在独立训练的保守模型中恢复出相同的类能量不变量，而在匹配的阻尼模型中未找到可比不变量。自主滚动时，该量会发生漂移。将隐态投影回其初始水平集可降低所有三个保守模型的滚动误差，而匹配的随机约束通常会增大误差。这些结果区分了具有动力学意义的不变量与仅可解码的关联量，揭示了一个具体失效模式：世界模型可从像素中学习物理约束，却在想象未来时违反该约束。
英文摘要
World models can predict video without learning dynamics that they reliably preserve. We test whether a frozen DreamerV3 trained only on pendulum video learns a scalar that its own latent transition treats as approximately conserved. A label-free search recovers the same energy-like invariant across independently trained conservative models, while the same procedure finds no comparable invariant in matched damped models. During autonomous rollouts, this quantity drifts. Projecting the latent state back toward its initial level set reduces rollout error in all three conservative models, whereas matched random constraints usually increase it. These results distinguish a dynamically meaningful invariant from a merely decodable correlate and reveal a concrete failure mode: a world model can learn a physical constraint from pixels yet violate that constraint when it imagines forward.
Comments10 pages, 5 figures. Code at https://github.com/Zarand3r/world-model-invariants
