具身可解释性：将因果理解与视觉-语言-动作模型的泛化联系起来
Embodied Interpretability: Linking Causal Understanding to Generalization in Vision-Language-Action Models
AI总结：
提出干预显著性评分（ISS）和干扰质量比（NMR），通过干预掩码估计视觉区域对动作预测的因果影响，并量化对任务无关特征的归因，实验表明NMR可预测泛化行为，ISS比现有方法提供更忠实的解释。
摘要：
视觉-语言-动作（VLA）策略在分布偏移下经常失败，这表明决策可能依赖于虚假的视觉相关性而非任务相关原因。我们将视觉-动作归因形式化为一个干预估计问题。据此，我们引入了干预显著性评分（ISS），一种用于估计视觉区域对动作预测的因果影响的干预掩码程序，以及干扰质量比（NMR），一种对任务无关特征归因的标量度量。我们分析了ISS的统计性质，表明它可以实现无偏估计，并刻画了动作预测误差为因果影响提供有效代理的条件。跨多种操作任务的实验表明，NMR预测泛化行为，且ISS比现有可解释性方法产生更忠实的解释。这些结果表明，干预归因为识别具身策略中的因果错位提供了一种简单的诊断方法。
英文摘要：
Vision-Language-Action (VLA) policies often fail under distribution shift, suggesting that decisions may depend on spurious visual correlations rather than task-relevant causes. We formulate visual-action attribution as an interventional estimation problem. Accordingly, we introduce the Interventional Significance Score (ISS), an interventional masking procedure for estimating the causal influence of visual regions on action predictions, and the Nuisance Mass Ratio (NMR), a scalar measure of attribution to task-irrelevant features. We analyze the statistical properties of ISS and show that it admits unbiased estimation, and we characterize conditions under which action prediction error provides a valid proxy for causal influence. Experiments across diverse manipulation tasks indicate that NMR predicts generalization behavior and that ISS yields more faithful explanations than existing interpretability methods. These results suggest that interventional attribution provides a simple diagnostic approach for identifying causal misalignment in embodied policies.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
