AI Paper Atlas

Research map / Graph learning / Graph algorithms and clustering

CastX: Cohort-Level Causal Inference Meets Statistical Testing for Faithful and Reliable GNN Explanations

Guanyuan Yu, Yijun Chen, Liang Xu, Gang Kou

AAAI 2026 · Graph learning › Graph algorithms and clustering · approach: Continuous DAG structure learning · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use cohort-level CATE instead of individual-level ITE (ablation)CastX · MutagACC-AUC: 0.93 -> 0.95Improved
use cohort-level CATE instead of individual-level ITE (ablation)CastX · VGACC-AUC: 0.79 -> 0.92Improved
use cohort-level CATE instead of individual-level ITE (ablation)CastX · REDDIT-5KACC-AUC: 0.52 -> 0.55Improved
use cohort-level CATE instead of individual-level ITE (ablation)CastX · VGACC@20%: 0.65 -> 0.79Improved
use cohort-level CATE instead of individual-level ITE (ablation)CastX · MutagCST: 0.30 -> 0.16Improved
use cohort-level CATE instead of individual-level ITE (ablation)CastX · Mutag, REDDIT-5K, VG under random confounder and data subset testsACC-AUC dropImproved
add permutation-test pruning of insignificant edgesCastX · Mutag example graphp-value based stopping: p = 0.076Improved
use CastX instead of baseline explainersGNN explainers · Mutag, REDDIT-5K, VGACC-AUC: average improvements of 8.96% and 13.33%Improved
use CastX instead of baseline explainersGNN explainers · REDDIT-5KCST: 0.24 vs Gem 0.15Hurt
use CastX instead of baseline explainersGNN explainers · VGrobustness to perturbation: drop of only 1.01% vs average 3.47% and 5.21%Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy