CausalGrade: Evidence-Guided Causal Graph Scoring with Jev
Abstract
Evaluating the structural quality of candidate causal graphs when the true causal structure is unknown is an important problem in applying causal discovery results. This paper proposes CausalGrade, a causal graph scoring method based on the probabilistic judgment model Jev. The method jointly provides Jev with evidence of association, conditional dependence, and directionality from observational data and edge probabilities produced by the causal discovery foundation model CDFM. Jev produces probabilistic judgments about direct causal edges, and a graph-level score is obtained by measuring the agreement between the adjacency structure of the graph under evaluation and these edge-level judgments. Across 300 synthetic datasets, the score exhibits a relatively strong negative correlation with structural error in all 20 combinations of four causal discovery algorithms and five variable counts, indicating that higher scores generally correspond to smaller structural errors. In 500 synthetic graph selection tasks, CausalGrade achieves a best-candidate hit rate of 54.4\%, exceeding the 48.4\% achieved by the Self-Compatibility metric κG, further supporting its ability to discriminate the structural quality of graphs under evaluation.
Related
- Causal graph rewriting
- Epidemiological Causal Graph Identification: Challenges, Identifiability and Algorithms
- Graph-Split Bayesian Causal Forest for Spatial Heterogeneous Treatment Effect Estimation
- Balance-Guided Forest for Causal Effect Estimation
- Causal DAG-Guided Reinforcement Learning for Long-Term Physical Activity Maintenance: Using Prior Data to Inform Future Deployment
BibTeX
@misc{zhong2026causalgrade,
title = {{CausalGrade: Evidence-Guided Causal Graph Scoring with Jev}},
author = {Jiangming Zhong and Jiawei Zhu and Guanting Wu and Qiaoxi Cai and Wanyi Chen and Jiale Li},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.40692},
eprint = {2610.40692},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.40692},
}
RIS
TY - UNPB
TI - CausalGrade: Evidence-Guided Causal Graph Scoring with Jev
AU - Jiangming Zhong
AU - Jiawei Zhu
AU - Guanting Wu
AU - Qiaoxi Cai
AU - Wanyi Chen
AU - Jiale Li
PY - 2026
DA - 2026/10/28
PB - arXiv
DO - 10.48550/arxiv.2610.40692
UR - https://arxiv.org/abs/2610.40692
AB - Evaluating the structural quality of candidate causal graphs when the true causal structure is unknown is an important problem in applying causal discovery results. This paper proposes CausalGrade, a causal graph scoring method based on the probabilistic judgment model Jev. The method jointly provides Jev with evidence of association, conditional dependence, and directionality from observational data and edge probabilities produced by the causal discovery foundation model CDFM. Jev produces probabilistic judgments about direct causal edges, and a graph-level score is obtained by measuring the agreement between the adjacency structure of the graph under evaluation and these edge-level judgments. Across 300 synthetic datasets, the score exhibits a relatively strong negative correlation with structural error in all 20 combinations of four causal discovery algorithms and five variable counts, indicating that higher scores generally correspond to smaller structural errors. In 500 synthetic graph selection tasks, CausalGrade achieves a best-candidate hit rate of 54.4\%, exceeding the 48.4\% achieved by the Self-Compatibility metric κ_G, further supporting its ability to discriminate the structural quality of graphs under evaluation.
ER -