你观察什么决定你如何识别因果效应：跨观测视图评估因果模型
What You Observe Determines How You Identify Causal Effects: Evaluating Causal Models across Observational Views
浏览论文内容
中文总结 AI 辅助
本研究提出CausalIDView基准，通过固定SCM和目标估计量仅改变观测视图，发现因果基础模型在不同识别机制下表现不稳定，而结合显式识别与预测模型的模块化方法更具竞争力。
中文摘要 AI 辅助
因果基础模型（CFMs）通过在由各种结构因果模型（SCMs）生成的数据上进行预训练，已被提出用于从观测数据中估计因果效应。然而，预训练环境和评估协议的差异使得难以评估其性能如何依赖于可用于因果识别的信息。为了实现受控比较，我们引入了CausalIDView，一个多视图基准，它固定SCM实现和目标估计量，仅改变提供给估计器的观测视图。在该基准维持的因果假设下，每个观测视图对应一个不同的识别机制。在这些匹配的视图上，没有CFM始终表现最佳，且模型排名变化显著。在受控结构变化下，CFMs表现出模型特定的失败，即当真实效应不变时无法维持稳定估计，以及无法跟踪真实的效应变化。我们还检验了将显式识别与强预测估计相结合是否有效。一种将预测性表格基础模型与特定机制识别程序配对的模块化方法，与CFMs相比具有竞争力，并优于其中几个。这些发现促使进行跨机制比较，以评估CFMs的经验价值。
英文摘要
Causal foundation models (CFMs) pre-trained on data generated from various structural causal models (SCMs) have been proposed for estimating causal effects from observational data. However, differences in pre-training environments and evaluation protocols make it difficult to assess how their performance depends on the information available for causal identification. To enable controlled comparisons, we introduce CausalIDView, a multi-view benchmark that holds fixed SCM realization and target estimand while varying only the observational view available to the estimator. Each observational view corresponds to a distinct identification regime under the benchmark's maintained causal assumptions. Across these matched views, no CFM consistently performs best and model rankings vary substantially. Under controlled structural changes, CFMs exhibit model-specific failures to maintain stable estimates when true effects are unchanged and to track genuine effect changes. We also examine whether combining explicit identification with strong predictive estimation is effective. A modular approach that pairs a predictive tabular foundation model with regime-specific identification procedures is competitive with CFMs and outperforms several of them. These findings motivate cross-regime comparisons to assess the empirical value of CFMs.
发表机构
- Yonsei University（延世大学）
- University of Illinois Urbana–Champaign（伊利诺伊大学厄巴纳-尚佩恩分校）
机构由 AI 辅助整理，请以论文原文为准。
