An automated LLM pipeline finds large differences in how well 11 real-world causal discovery benchmarks align with recent domain literature.
arXiv preprint arXiv:2512.11219 , year=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Provides necessary and sufficient conditions for ATE identifiability under selection bias by characterizing propensity and selection probabilities via weak assumptions on probability classes.
citing papers explorer
-
Consistency evaluation of benchmarks used for causal discovery
An automated LLM pipeline finds large differences in how well 11 real-world causal discovery benchmarks align with recent domain literature.
-
Towards a holistic understanding of Selection Bias for Causal Effect Identification
Provides necessary and sufficient conditions for ATE identifiability under selection bias by characterizing propensity and selection probabilities via weak assumptions on probability classes.