Graph2Video: Leveraging Video Models to Model Dynamic Graph Evolution

2026-03-09

AI Summary

This work proposes a novel approach to dynamic graph learning by drawing inspiration from video modeling. Existing methods struggle to effectively capture fine-grained temporal interactions, long-range dependencies, and link-specific relational dynamics in temporal link prediction. To address these limitations, the authors formulate the temporal neighborhood of a target link as a “graph video” and leverage the spatiotemporal inductive biases inherent in foundational video models. This enables unified modeling of both local interaction details and long-term evolutionary patterns, yielding lightweight, plug-and-play link-level embeddings. Extensive experiments on multiple benchmark datasets demonstrate that the proposed method significantly outperforms current state-of-the-art approaches, thereby validating the effectiveness of cross-domain transfer of spatiotemporal modeling capabilities to dynamic graph representation learning.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Planning, Routing, and Scheduling: Temporal Planning

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Dynamic graphs are common in real-world systems such as social media, recommender systems, and traffic networks. Existing dynamic graph models for link prediction often fall short in capturing the complexity of temporal evolution. They tend to overlook fine-grained variations in temporal interaction order, struggle with dependencies that span long time horizons, and offer limited capability to model pair-specific relational dynamics. To address these challenges, we propose \textbf{Graph2Video}, a video-inspired framework that views the temporal neighborhood of a target link as a sequence of "graph frames". By stacking temporally ordered subgraph frames into a "graph video", Graph2Video leverages the inductive biases of video foundation models to capture both fine-grained local variations and long-range temporal dynamics. It generates a link-level embedding that serves as a lightweight and plug-and-play link-centric memory unit. This embedding integrates seamlessly into existing dynamic graph encoders, effectively addressing the limitations of prior approaches. Extensive experiments on benchmark datasets show that Graph2Video outperforms state-of-the-art baselines on the link prediction task in most cases. The results highlight the potential of borrowing spatio-temporal modeling techniques from computer vision as a promising and effective approach for advancing dynamic graph learning.

Problem

• dynamic graphs
• link prediction
• temporal evolution
• graph representation learning
• spatio-temporal modeling

Innovation

• dynamic graph
• video foundation models
• link prediction
• temporal modeling
• graph representation learning