Net-Ev$^2$: A Generative Simulator for Network Event Evolution

2026-06-10

AI Summary

This work addresses the challenge that existing methods struggle to jointly model the structured attributes and unstructured semantics of network disruption events while preserving topological consistency in simulations. To overcome this limitation, the authors propose Net-Ev², a generative simulation framework that uniquely integrates event semantics with network topology. By leveraging structure-guided masked pretraining and a topology-aware diffusion process, Net-Ev² generates topology-preserving event evolution sequences from natural language inputs alone. Key innovations include a U-Net–inspired graph downsampling/upsampling mechanism, a multimodal benchmark dataset (Net-Ev²-6.5M), and JL-MMD, a novel metric for evaluating topological fidelity. Experiments demonstrate state-of-the-art performance across multiple large-scale road network datasets, achieving both strong generalization and high topological accuracy.

Abstract

Reducing real-world trial and error has long been a central goal of decision making, and generative simulators advance this goal by modeling the evolution of future states. An even more challenging yet meaningful task is simulating how disturbance events (e.g., accidents) propagate their impacts across real-world networks. The existing approaches fall short of modeling both structured attributes and unstructured semantics of events, and capturing topological structures in simulating network event evolution. Therefore, we are motivated to propose Net-Ev$^2$ ($\underline{\textbf{Net}}$work $\underline{\textbf{Ev}}$ent $\underline{\textbf{Ev}}$olution), a novel generative simulator that jointly leverages event cues while preserving network topology in simulations. Specifically, the framework consists of two stages, namely structure-guided masked pre-training and topology-aware diffusion process, which is achieved by U-Net-like graph downsampling and upsampling during denoising. At inference time, Net-Ev$^2$ can generate simulations using natural-language event input only, with greater flexibility for practical usage. Furthermore, we introduce Net-Ev$^2$-6.5M, a multimodal benchmark of aligned event and network traffic data across four large-scale road networks, as well as a new topology-aware metric, namely JL-MMD, to evaluate topological fidelity in generated network dynamics. Extensive experiments demonstrate the state-of-the-art performance and strong generalization ability of Net-Ev$^2$. Code is made available at https://github.com/Guangyu4/Net-Ev-2.

Problem

• network event evolution
• generative simulation
• topological structure
• disturbance propagation
• multimodal event modeling

Innovation

• generative simulation
• network event evolution
• topology-aware diffusion
• multimodal benchmark
• graph denoising