Graph2Video: Leveraging Video Models to Model Dynamic Graph Evolution

TL;DR

Graph2Video introduces a novel video-inspired framework that models dynamic graph evolution by capturing fine-grained and long-range temporal dependencies, significantly improving link prediction performance.

Contribution

It proposes a new approach that leverages video foundation models to better capture temporal dynamics in evolving graphs, addressing limitations of existing methods.

Findings

Outperforms state-of-the-art baselines on benchmark datasets

Effectively captures both local variations and long-term dependencies

Demonstrates the potential of computer vision techniques in graph learning

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.