Node Role-Guided LLMs for Dynamic Graph Clustering

2026-03-14

AI Summary

This work addresses the limited interpretability of dynamic graph clustering, which hinders its deployment in safety-critical domains such as healthcare and transportation where semantically meaningful evolutionary explanations are essential. The authors propose the first end-to-end interpretable framework that disentangles node roles from cluster structures via orthogonal subspaces and introduces five learnable semantic role prototypes—Leader, Contributor, Wanderer, Connector, and Newcomer—to guide large language models (LLMs) in hierarchical reasoning. This approach simultaneously produces clustering results and natural language explanations, uniquely integrating role-based semantic prototypes with LLMs to enable consistency-aware refinement under weak supervision. Evaluated on four synthetic and six real-world dynamic graph benchmarks, the method significantly outperforms existing black-box models, achieving notable advances in effectiveness, robustness, and interpretability.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs 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: Architectures and workflows that use LLMs for crowd work

Abstract

Dynamic graph clustering aims to detect and track time-varying clusters in dynamic graphs, revealing how complex real-world systems evolve over time. However, existing methods are predominantly black-box models. They lack interpretability in their clustering decisions and fail to provide semantic explanations of why clusters form or how they evolve, severely limiting their use in safety-critical domains such as healthcare or transportation. To address these limitations, we propose an end-to-end interpretable framework that maps continuous graph embeddings into discrete semantic concepts through learnable prototypes. Specifically, we first decompose node representations into orthogonal role and clustering subspaces, so that nodes with similar roles (e.g., hubs, bridges) but different cluster affiliations can be properly distinguished. We then introduce five node role prototypes (Leader, Contributor, Wanderer, Connector, Newcomer) in the role subspace as semantic anchors, transforming continuous embeddings into discrete concepts to facilitate LLM understanding of node roles within communities. Finally, we design a hierarchical LLM reasoning mechanism to generate both clustering results and natural language explanations, while providing consistency feedback as weak supervision to refine node representations. Experimental results on four synthetic and six real-world benchmarks demonstrate the effectiveness, interpretability, and robustness of DyG-RoLLM. Code is available at https://github.com/Clearloveyuan/DyG-RoLLM.

Problem

• dynamic graph clustering
• interpretability
• semantic explanation
• node roles
• black-box models

Innovation

• interpretable clustering
• node role prototypes
• dynamic graph representation
• large language models (LLMs)
• orthogonal subspace decomposition