Event-Aware Prompt Learning for Dynamic Graphs

2025-10-13

AI Summary

Existing dynamic graph learning methods commonly neglect the influence of historical interaction events on node evolution, thereby limiting their temporal modeling capability. To address this, we propose Event-aware Prompt learning (EVP), the first framework to explicitly incorporate historical interaction events into dynamic graph prompt learning. EVP achieves effective alignment between event knowledge and downstream tasks through fine-grained event feature adaptation and cross-timestep aggregation. Importantly, EVP is plug-and-play: it enhances state-of-the-art dynamic graph neural networks without modifying their backbone architectures. Extensive experiments on four standard benchmark datasets demonstrate that EVP consistently improves performance—particularly in link prediction—validating both the efficacy of explicit historical event modeling and the generality and transferability of the proposed framework.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Models for Web evolution

Abstract

Real-world graph typically evolve via a series of events, modeling dynamic interactions between objects across various domains. For dynamic graph learning, dynamic graph neural networks (DGNNs) have emerged as popular solutions. Recently, prompt learning methods have been explored on dynamic graphs. However, existing methods generally focus on capturing the relationship between nodes and time, while overlooking the impact of historical events. In this paper, we propose EVP, an event-aware dynamic graph prompt learning framework that can serve as a plug-in to existing methods, enhancing their ability to leverage historical events knowledge. First, we extract a series of historical events for each node and introduce an event adaptation mechanism to align the fine-grained characteristics of these events with downstream tasks. Second, we propose an event aggregation mechanism to effectively integrate historical knowledge into node representations. Finally, we conduct extensive experiments on four public datasets to evaluate and analyze EVP.

Problem

• Modeling dynamic graph interactions while incorporating historical event knowledge
• Enhancing existing methods with event-aware prompt learning framework
• Aligning historical event characteristics with downstream task requirements

Innovation

• Event-aware prompt learning for dynamic graphs
• Event adaptation mechanism aligns event characteristics
• Event aggregation integrates historical knowledge into nodes