The paper introduces a logic-based framework that extracts global logical rules for node classification in Simple Graph Convolution (SGC) networks. It uses minimal abductive explanations—small sets of node-feature pairs that preserve a node’s predicted class—as an intermediate step. Decision trees trained on these explanations yield compact global rules that retain high fidelity to the original SGC model, as demonstrated on benchmark datasets.
arXiv:2607. 04600v1 Announce Type: new Abstract: Graph eXplainable AI (G-XAI) is increasingly important for making Graph Neural Networks interpretable and accountable.
arXiv:2605. 30747v2 Announce Type: replace Abstract: Logical rules constitute a cornerstone of knowledge graph (KG) reasoning, valued for their interpretability and ability to model relational patterns.
SLogic introduces a subgraph-informed approach to logical rule learning for knowledge graph completion, assigning query-dependent scores to rules instead of a single global weight. The framework uses a context-aware scoring function that evaluates the importance of a rule based on the local subgraph around the query’s head entity, aligning with the specificity principle of commonsense reasoning. Experiments on benchmark datasets demonstrate that SLogic performs competitively with other rule-based methods while producing human-readable, query-specific explanations.
Graph eXplainable AI (G-XAI) is increasingly important for making Graph Neural Networks interpretable and accountable. While a growing number of explainers are available, choosing the right method and assessing the trustworthiness of its outputs remains unclear.
arXiv:2511. 11046v3 Announce Type: replace-cross Abstract: Graph neural networks (GNNs) have become an indispensable tool for analyzing relational data.
arXiv:2512. 01113v2 Announce Type: replace-cross Abstract: Algorithmic reasoning -- the ability to perform step-by-step logical inference -- is a synthetic benchmark for evaluating multi-step reasoning abilities, designed for graph neural networks and also for transformer models.
The paper introduces a flexible symbolic framework that efficiently computes logical explanations for deep neural networks by parameterizing explanations with internal neuron activations and leveraging general-purpose logical engines like SMT solvers. Unlike previous methods that rely on specialized verifiers or are limited to individual input features, this approach is not restricted in shape and can scale to deep architectures. Experiments on image recognition and medical benchmarks demonstrate improved computational efficiency and the ability to explain networks that were previously intractable for logic-based methods.
The paper presents a comparative analysis of six state‑of‑the‑art counterfactual explainers for graph neural networks, focusing on methods that can both add and remove edges to alter model predictions. It evaluates these explainers across diverse real‑world and synthetic datasets, covering binary and multi‑class graph and node classification tasks, using a range of quantitative and qualitative metrics. The study highlights the trade‑offs between explanation size, coverage, and quality, aiming to pinpoint each method’s strengths and weaknesses to inform future research.
arXiv:2606. 17882v1 Announce Type: new Abstract: Bridges between graph neural networks (GNNs) and logical formalisms have been established by fixing architectural choices, such as the types of aggregation, combination, and activation functions.
The paper introduces SLM-Conditioned Hierarchical Relation Routing, a graph neural network architecture that incorporates a small language model to guide message selection in labeled property graphs. It combines structural states, node and relationship property encodings, and relationship types into messages, then uses a parameter‑efficient language model to generate target‑conditioned routing queries that filter relevant messages and route information across relation‑level summaries. The resulting representation updates a topology anchor with bounded residuals, preserving structural evidence while allowing contextual semantic information to influence predictions, and offers interpretable analysis at both neighbor and relationship‑type levels.
arXiv:2607. 21381v1 Announce Type: new Abstract: Instance-level explanations aim to reveal the rationale behind a model's decisions for a specific graph.
arXiv:2608. 14121v1 Announce Type: cross Abstract: Graph Neural Networks (GNNs) can solve prediction tasks by unintentionally exploiting shortcuts---that is, edges, nodes, and features that correlate with but are not causal for the prediction---which compromise their reliability in out-of-distribution tasks.