cs.LGFeb 20, 2026

Causal Neighbourhood Learning for Invariant Graph Representations

Authors: Simi Job, Xiaohui Tao, Taotao Cai, Haoran Xie, Jianming Yong

Organizations: School of Science, Engineering and Digital Technologies, University of Southern Queensland, Australia. · School of Data Science, Lingnan University, Hong Kong.

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Graph data often contain noisy and spurious correlations that mask the true causal relationships, which are essential for enabling graph models to make predictions based on the underlying causal structure of the data. Dependence on spurious connections makes it challenging for traditional Graph Neural Networks (GNNs) to generalize effectively across different graphs. Furthermore, traditional aggregation methods tend to amplify these spurious patterns, limiting model robustness under distribution shifts. To address these issues, we propose Causal Neighbourhood Learning with Graph Neural Networks (CNL-GNN), a novel framework that performs causal interventions on graph structure. CNL-GNN effectively identifies and preserves causally relevant connections and reduces spurious influences through the generation of counterfactual neighbourhoods and adaptive edge perturbation guided by learnable importance masking and an attention-based mechanism. In addition, by combining structural-level interventions with the disentanglement of causal features from confounding factors, the model learns invariant node representations that are robust and generalize well across different graph structures. Our approach improves causal graph learning beyond traditional feature-based methods, resulting in a robust classification model. Extensive experiments on four publicly available datasets, including multiple domain variants of one dataset, demonstrate that CNL-GNN outperforms state-of-the-art GNN models.

Figures & tables

Fig. 1: Illustration of causal concepts using SCMs. (a) cause C influences effect E . (b) sets E to a fixed value. (c) modifies function or noise affecting E . ( NEN{{}_{E}} : noise) (d) updates noise to evaluate alternative outcomes without changing causal structure. (T=1, B=1 : observed treatment and outcome NB=1: inferred latent condition)
Fig. 2: CNL-GNN Architecture: (i) Enforces structure-invariant representations by perturbing neighbourhoods with dissimilar neighbours. (ii) Estimates edge relevance using attention to prioritize causal edges and guide structural perturbations. (iii) Selectively perturbs inter-group edges and masks low-importance edges to enhance causal robustness. (iv) Disentangles and fuses causal and non-causal features using gating for robust prediction.
Dataset# Nodes# Edges# Classes
Cora2708105567
Citeseer332791046
PubMed19717886483
Twitch-DE94983157742
Twitch-EN7126353242
Twitch-PT1912312992
TABLE I: Overview of Datasets used in the Study
Fig. 3: Precision, Recall and F1 evaluation under distribution shifts across Twitch domains. Arrows indicate relative F1 drop compared to the DE domain.
ModelTwitchCoraCiteseerPubMed
GCN-ICL [ 26 ]95.9873.9442.7177.91
GAT-ICL [ 26 ]96.7174.4751.7684.15
ACE-GCN [ 27 ]64.5061.9035.9373.03
ACE-GAT [ 27 ]68.1272.3758.8878.67
CaNet [ 45 ]66.1784.3273.7087.19
CIE [ 49 ]38.0883.3667.3484.71
TABLE II: F-Scores for node classification tasks
Fig. 4: F1 Scores across datasets (CNL-GNN vs. Ablated Variants). CNL-GNN bars are shown in blue, ablations are white with hatch patterns.
Fig. 5: Sensitivity Analysis: Δ\Delta F1-score vs. Edge Drop Rate for Cora , Citeseer and Pubmed . Values indicate deviation from baseline at drop rate = 0.1.
More in this topic

Graph Representation Learning

Graph representation learning develops methods for encoding nodes, edges, and whole graphs for prediction, comparison, and transfer.

See all 292 papers

Explore similar work

CardsList
Feb 20, 2026

Optimizing Graph Causal Classification Models: Estimating Causal Effects and Addressing Confounders

Graph data is becoming increasingly prevalent due to the growing demand for relational insights in AI across various domains. Organizations regularly use graph data to solve complex problems involving relationships and connections. Causal learning is especially important in this context, since it helps to understand cause-effect relationships rather than mere associations. Since many real-world systems are inherently causal, graphs can efficiently model these systems. However, traditional graph machine learning methods including graph neural networks (GNNs), rely on correlations and are sensitive to spurious patterns and distribution changes. On the other hand, causal models enable robust predictions by isolating true causal factors, thus making them more stable under such shifts. Causal learning also helps in identifying and adjusting for confounders, ensuring that predictions reflect true causal relationships and remain accurate even under interventions. To address these challenges and build models that are robust and causally informed, we propose CCAGNN, a Confounder-Aware causal GNN framework that incorporates causal reasoning into graph learning, supporting counterfactual reasoning and providing reliable predictions in real-world settings. Comprehensive experiments on six publicly available datasets from diverse domains show that CCAGNN consistently outperforms leading state-of-the-art models.

Simi Job, Xiaohui Tao, Taotao Cai +3

School of Mathematics, Physics, and Computing, University of Southern Queensland, Australia · School of Data Science, Lingnan University, Hong Kong · School of Business, University of Southern Queensland, Australia +1

Graph Neural NetworksCausal InferenceCounterfactual Learning
AskPDF
Apr 10, 2026

A Closer Look at the Application of Causal Inference in Graph Representation Learning

Modeling causal relationships in graph representation learning remains a fundamental challenge. Existing approaches often draw on theories and methods from causal inference to identify causal subgraphs or mitigate confounders. However, due to the inherent complexity of graph-structured data, these approaches frequently aggregate diverse graph elements into single causal variables, an operation that risks violating the core assumptions of causal inference. In this work, we prove that such aggregation compromises causal validity. Building on this conclusion, we propose a theoretical model grounded in the smallest indivisible units of graph data to ensure that the causal validity is guaranteed. With this model, we further analyze the costs of achieving precise causal modeling in graph representation learning and identify the conditions under which the problem can be simplified. To empirically support our theory, we construct a controllable synthetic dataset that reflects realworld causal structures and conduct extensive experiments for validation. Finally, we develop a causal modeling enhancement module that can be seamlessly integrated into existing graph learning pipelines, and we demonstrate its effectiveness through comprehensive comparative experiments.

Hang Gao, Kunyu Li, Huang Hong +2

Institute of Software Chinese Academy of Science

Graph Representation LearningCausal Inference
AskPDF
Apr 21, 2026

Inductive Subgraphs as Shortcuts: Causal Disentanglement for Heterophilic Graph Learning

Heterophily is a prevalent property of real-world graphs and is well known to impair the performance of homophilic Graph Neural Networks (GNNs). Prior work has attempted to adapt GNNs to heterophilic graphs through non-local neighbor extension or architecture refinement. However, the fundamental reasons behind misclassifications remain poorly understood. In this work, we take a novel perspective by examining recurring inductive subgraphs, empirically and theoretically showing that they act as spurious shortcuts that mislead GNNs and reinforce non-causal correlations in heterophilic graphs. To address this, we adopt a causal inference perspective to analyze and correct the biased learning behavior induced by shortcut inductive subgraphs. We propose a debiased causal graph that explicitly blocks confounding and spillover paths responsible for these shortcuts. Guided by this causal graph, we introduce Causal Disentangled GNN (CD-GNN), a principled framework that disentangles spurious inductive subgraphs from true causal subgraphs by explicitly blocking non-causal paths. By focusing on genuine causal signals, CD-GNN substantially improves the robustness and accuracy of node classification in heterophilic graphs. Extensive experiments on real-world datasets not only validate our theoretical findings but also demonstrate that our proposed CD-GNN outperforms state-of-the-art heterophily-aware baselines.

Xiangmeng Wang, Qian Li, Haiyang Xia +3

Curtin University Perth, Australia · The Education University of Hong Kong New Territories, Hong Kong

Disentangled Representation LearningGraph Neural NetworksCausal Representation LearningGraph Heterophily
AskPDF