MobiWorld: World Models for Mobile Wireless Network

arXiv 25.7 2025 61.3 method, application

TLDR

MobiWorld is a generative world model using diffusion models for high-fidelity simulation of mobile wireless networks to support planning and optimization.

Reasoning

The paper presents a novel generative world model for mobile networks, integrating heterogeneous data and enabling controllable simulation. Its strength lies in the explicit use of diffusion models for joint distribution modeling, but the abstract lacks details on real-world validation and does not address video or interactive aspects directly.

Read-first score

Read-first score 61.3, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 53.

Recency 8%

86.7

Uses a gentle age decay so recent papers surface without erasing older foundations. 2025

Topical relevance 42%

75.7

Uses existing LLM keyword relevance scores normalized to 0-100. world model,world simulator,generative world model,interactive world model,video world model,world dynamics prediction,model-based reinforcement learning world model

Methodology quality 25%

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=experiment,result

Reproducibility 25%

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=none

Field roles

Frontier

Rank sensitivity

Stability: volatile; rank range: 375.

Keyword Scores

world model

generative world model

world simulator

interactive world model

world dynamics prediction

model-based reinforcement learning world model

video world model

Deep Analysis

Innovations

Integration of heterogeneous data sources (sensors, mobile devices, base stations) and multimodal data types (sequences, images) for mobile network simulation.
Capability to generate both network element-level observations (e.g., traffic load, user distribution) and system-level performance indicators (e.g., throughput, energy consumption).
Built upon advanced diffusion models for controllable generation by modeling the joint distribution between mobile network data and diverse conditional factors (spatiotemporal contexts, user behaviors, optimization policies).

Methodology

MobiWorld is a generative world model built upon advanced diffusion models. It integrates heterogeneous data sources and multimodal data types, and models the joint distribution between mobile network data and diverse conditional factors including spatiotemporal contexts, user behaviors, and optimization policies to enable controllable generation of dynamic network states.

Key Results

In a collaborative energy-saving scenario, an agent uses observations and rewards generated by MobiWorld to optimize base station sleep and user offloading policies. Experimental results show that MobiWorld exhibits strong controllable generation performance and outperforms traditional methods in energy optimization.

Resources

Paper