AI 中文总结
本文提出IMPLY方法，通过反转模拟器读取推演隐含物理属性，并以校准推动为锚点评估世界模型推演的一致性，解决自一致性无法区分正确与错误物体理解的问题。
AI 中文摘要
一个世界模型被问及如果物体以不同速度被推动会发生什么时，会产生多个未来情景。如果模型对物体有正确的理解，这些未来情景在关于该物体的方面是一致的：每个情景都暗示相同的质量和摩擦力。目前用于审查世界-动作模型的一致性检查，仅询问模型的未来情景是否彼此一致，而这些检查都不了解任何物理知识。我们表明这还不够，并提出了替代方法。IMPLY通过反转模拟器来读取每个推演所隐含的物理属性，并根据一个物体能否解释所有推演来对一组推演进行评分，该评分锚定于模型已观察到的两次校准推动。在受控环境中，自一致性会给一个忽略物体并总是预测典型推动的模型打满分；而锚定方法则能揭露这一问题（AUROC 0.70 对比 1.00）。在真实模型上，即适应了场景的V-JEPA 2-AC，同样的情况也会发生。给定其自身的校准推动，模型能跟踪物体（每个物体与真实值的相关性为0.91）；而给定另一个物体的校准推动，则不能（相关性为0.05）。自一致性无法区分这两种情况，在52%的物体上偏好正确的证据，这相当于随机水平；而锚定不一致性在73%的物体上偏好正确的证据，并且与推演误差的相关性为0.92-0.99。当用于在候选推演集中进行选择时，它的表现与能看见真实情况的预言机相差在0.003以内。一个内化了错误物体的模型，其自一致性与内化了正确物体的模型完全相同；一致性必须锚定于证据。
英文摘要
A world model asked what happens if an object is pushed at several speeds produces several futures. If the model has the object in mind, those futures agree about it: each implies the same mass and friction. The consistency checks now used to vet world-action models ask whether a model's futures agree with each other, and none of them knows any physics. We show that this is not enough, and what to do instead. IMPLY reads the physics each rollout implies by inverting a simulator and scores a set of rollouts by how well one object explains all of them, anchored to two calibration pushes the model has observed. In a controlled setting, self-consistency gives a perfect score to a model that ignores the object and always predicts a typical push; anchoring exposes it (AUROC 0.70 versus 1.00). On a real model, V-JEPA 2-AC adapted to the scene, the same thing happens. Given its own calibration pushes the model tracks the object (per-object correlation with the truth 0.91); given another object's, it does not (0.05). Self-consistency cannot tell these apart, preferring the right evidence on 52% of objects, chance level, while anchored disagreement prefers it on 73% and correlates 0.92-0.99 with the rollouts' error. Used to choose among candidate rollout sets, it comes within 0.003 of an oracle that sees the truth. A model that has internalised the wrong object is exactly as self-consistent as one that has internalised the right one; consistency has to be anchored to evidence.
Comments7 pages, 2 figures, 3 tables
