AI 中文总结
本研究提出噪声耦合双回滚框架，解决交互式视频世界模型的反事实生成问题，规避近似逆问题，定义时空局部性度量，待开展实验验证。
AI 中文摘要
交互式视频世界模型会在动作流下自回归地生成回滚序列，但它们的训练与评估几乎仅针对事实预测。我们研究回滚过程中的反事实生成：给定模型自身生成的轨迹，若从步骤t*起动作与原轨迹不同，将会发生什么？我们提出噪声耦合双回滚——事实分支与反事实分支共享已生成的前缀及未来的外部噪声序列，仅在干预点处的动作流上产生分歧。由于事实分支是模型自身生成的，其外部噪声完全已知：Pearl反事实推理流程中的溯因步骤在构造上是精确的，规避了基于编辑的方法所面临的近似逆问题。噪声耦合进一步将最小变化原则转化为可针对单一样本验证的性质：我们定义了一种时空局部性度量，该度量会惩罚干预的因果后代之外的分歧，可对照模拟器真实值计算，无需学习评判器。在t*处分叉模拟器状态可得到真实的反事实重渲染，我们将其用作训练后的可验证奖励。本研究确立了形式框架、度量定义与定位，实验内容待后续开展。
英文摘要
Interactive video world models generate rollouts autoregressively under an action stream, yet they are trained and evaluated almost exclusively on factual prediction. We study counterfactual generation inside the rollout: given a trajectory the model has itself generated, what would have happened had the actions differed from step t* onward? We formalize noise-coupled twin rollouts --- a factual and a counterfactual branch sharing the generated prefix and the future exogenous noise sequence, diverging only in the action stream at an intervention point. Because the factual branch is self-generated, its exogenous noise is known exactly: the abduction step of Pearl's counterfactual procedure is exact by construction, sidestepping the approximate-inversion problem faced by editing-based pipelines. Noise coupling further turns the minimal-change principle into a per-sample verifiable property: we define a spatiotemporal locality metric that penalizes divergence outside the causal descendants of the intervention, computable against simulator ground truth without a learned judge. Forking the simulator state at t* yields ground-truth counterfactual re-renders, which we use as verifiable rewards for post-training. This note establishes the formal framework, metric definitions, and positioning; experiments are forthcoming.
