The Final-Stage Bottleneck: A Systematic Dissection of the R-Learner for Network Causal Inference

ArXi:2511.13018v3 Announce Type: replace The R-Learner is a powerful, theoretically-grounded framework for estimating heterogeneous treatment effects, prized for its robustness to nuisance model errors. However, its application to network data, where causal heterogeneity is often graph-dependent, presents a critical challenge to its core assumption of a well-specified final-stage model. In this paper, we conduct a large-scale empirical study to systematically dissect the R-Learner framework on graphs.