Neighbourhood Transformer: Switchable Attention for Monophily-Aware Graph Learning

2026-04-10

AI Summary

This work addresses the performance limitations of conventional graph neural networks (GNNs) on heterogeneous graphs, which stem from their reliance on the homophily assumption. To overcome this, the authors propose Neighborhood Transformer, a novel framework that explicitly models monophily—a tendency for nodes to connect with others sharing similar attributes—by replacing standard message passing with a local self-attention mechanism within each node’s neighborhood. The method further incorporates a switchable attention scheme and adaptive neighborhood partitioning strategy, significantly reducing computational overhead while preserving expressive power. Evaluated across ten real-world datasets—including five heterogeneous graphs—the proposed approach consistently outperforms state-of-the-art models, achieving over 95% reduction in memory consumption and up to 92.67% decrease in runtime.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

Graph neural networks (GNNs) have been widely adopted in engineering applications such as social network analysis, chemical research and computer vision. However, their efficacy is severely compromised by the inherent homophily assumption, which fails to hold for heterophilic graphs where dissimilar nodes are frequently connected. To address this fundamental limitation in graph learning, we first draw inspiration from the recently discovered monophily property of real-world graphs, and propose Neighbourhood Transformers (NT), a novel paradigm that applies self-attention within every local neighbourhood instead of aggregating messages to the central node as in conventional message-passing GNNs. This design makes NT inherently monophily-aware and theoretically guarantees its expressiveness is no weaker than traditional message-passing frameworks. For practical engineering deployment, we further develop a neighbourhood partitioning strategy equipped with switchable attentions, which reduces the space consumption of NT by over 95% and time consumption by up to 92.67%, significantly expanding its applicability to larger graphs. Extensive experiments on 10 real-world datasets (5 heterophilic and 5 homophilic graphs) show that NT outperforms all current state-of-the-art methods on node classification tasks, demonstrating its superior performance and cross-domain adaptability. The full implementation code of this work is publicly available at https://github.com/cf020031308/MoNT to facilitate reproducibility and industrial adoption.

Problem

• heterophilic graphs
• homophily assumption
• graph neural networks
• monophily

Innovation

• Neighbourhood Transformer
• monophily-aware
• switchable attention
• heterophilic graphs
• graph neural networks