Abstract
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.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Jung, H., Seo, G., Byun, H., Lee, J., & Song, K. (2026). What You Observe Determines How You Identify Causal Effects: Evaluating Causal Models across Observational Views. https://omanscience.com/en/articles/what-you-observe-determines-how-you-identify-causal-effects-evaluating-causal-models-across-observational-views
MLA 9
Jung, Heejin, et al. "What You Observe Determines How You Identify Causal Effects: Evaluating Causal Models across Observational Views." https://omanscience.com/en/articles/what-you-observe-determines-how-you-identify-causal-effects-evaluating-causal-models-across-observational-views.
Chicago (author–date)
Jung, Heejin, Gyeongdeok Seo, Hoyoon Byun, Joseph Lee, and Kyungwoo Song. 2026. "What You Observe Determines How You Identify Causal Effects: Evaluating Causal Models across Observational Views." https://omanscience.com/en/articles/what-you-observe-determines-how-you-identify-causal-effects-evaluating-causal-models-across-observational-views.
Harvard
Jung, H., Seo, G., Byun, H., Lee, J. and Song, K. (2026) 'What You Observe Determines How You Identify Causal Effects: Evaluating Causal Models across Observational Views', Available at: https://omanscience.com/en/articles/what-you-observe-determines-how-you-identify-causal-effects-evaluating-causal-models-across-observational-views.
Vancouver
Jung H, Seo G, Byun H, Lee J, Song K. What You Observe Determines How You Identify Causal Effects: Evaluating Causal Models across Observational Views. https://omanscience.com/en/articles/what-you-observe-determines-how-you-identify-causal-effects-evaluating-causal-models-across-observational-views
IEEE
H. Jung, G. Seo, H. Byun, J. Lee, and K. Song, "What You Observe Determines How You Identify Causal Effects: Evaluating Causal Models across Observational Views," https://omanscience.com/en/articles/what-you-observe-determines-how-you-identify-causal-effects-evaluating-causal-models-across-observational-views.