通过干预式输入分析，揭示时间模型依赖的因果结构。
Causal-INSIGHT: Probing Temporal Models to Extract Causal Structure
- 在推理时施加干预式输入钳制，探测模型响应。
- 可准确识别时间滞后影响关系，提升延迟定位精度。
- 适用于多种模型架构，无需真实图谱标签。
理解多变量时间序列中的有向时间交互对解释复杂动力系统及其预测模型至关重要。我们提出Causal-INSIGHT，一种模型无关、事后分析的解释框架，用于从训练好的时间预测器中提取模型隐含的（预测因子依赖的）、有向的、时间滞后的影响力结构。该方法不试图推断数据生成过程的因果结构，而是分析固定预训练预测器在推理时对系统性干预式输入钳制的响应。基于这些响应，构建反映预测器依赖关系的有向时间影响信号，并引入Qbic这一稀疏感知图选择准则，在不依赖真实图标签的前提下，平衡预测保真度与结构复杂度。在合成、模拟和真实基准上的实验表明，Causal-INSIGHT可泛化至多种主干架构，保持竞争性结构准确性，并显著提升现有预测器的时间延迟定位性能。
原文摘要 · Abstract (English)
Understanding directed temporal interactions in multivariate time series is essential for interpreting complex dynamical systems and the predictive models trained on them. We present Causal-INSIGHT, a model-agnostic, post-hoc interpretation framework for extracting model-implied (predictor-dependent), directed, time-lagged influence structure from trained temporal predictors. Rather than inferring causal structure at the level of the data-generating process, Causal-INSIGHT analyzes how a fixed, pre-trained predictor responds to systematic, intervention-inspired input clamping applied at inference time. From these responses, we construct directed temporal influence signals that reflect the dependencies the predictor relies on for prediction, and introduce Qbic, a sparsity-aware graph selection criterion that balances predictive fidelity and structural complexity without requiring ground-truth graph labels. Experiments across synthetic, simulated, and realistic benchmarks show that Causal-INSIGHT generalizes across diverse backbone architectures, maintains competitive structural accuracy, and yields significant improvements in temporal delay localization when applied to existing predictors.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。