Edge Large AI Models: Revolutionizing 6G Networks

2025-05-01

AI Summary

To address the core challenges of deploying large AI models (LAMs) in resource-constrained wireless networks—namely, deployment difficulty, high communication overhead, and poor generalization for 6G edge intelligence—this work proposes a novel collaborative fine-tuning and full-parameter training framework. It establishes an integrated architecture encompassing model decomposition, distributed training, wireless-aware scheduling, and microservice-based inference. Crucially, it pioneers the deep integration of edge-deployed LAMs into the physical layer design, enabling AI-native channel prediction and adaptive beamforming. Experimental evaluation under representative 6G scenarios demonstrates: >40% improvement in channel prediction accuracy, 35% increase in beam selection accuracy, and significant reductions in both communication overhead and end-to-end latency. These results validate the feasibility and effectiveness of edge LAMs in real-world wireless environments.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Planning, Routing, and Scheduling: Planning with Language Models
• Natural Language Processing: (Large) Language Models

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Large artificial intelligence models (LAMs) possess human-like abilities to solve a wide range of real-world problems, exemplifying the potential of experts in various domains and modalities. By leveraging the communication and computation capabilities of geographically dispersed edge devices, edge LAM emerges as an enabling technology to empower the delivery of various real-time intelligent services in 6G. Unlike traditional edge artificial intelligence (AI) that primarily supports a single task using small models, edge LAM is featured by the need of the decomposition and distributed deployment of large models, and the ability to support highly generalized and diverse tasks. However, due to limited communication, computation, and storage resources over wireless networks, the vast number of trainable neurons and the substantial communication overhead pose a formidable hurdle to the practical deployment of edge LAMs. In this paper, we investigate the opportunities and challenges of edge LAMs from the perspectives of model decomposition and resource management. Specifically, we propose collaborative fine-tuning and full-parameter training frameworks, alongside a microservice-assisted inference architecture, to enhance the deployment of edge LAM over wireless networks. Additionally, we investigate the application of edge LAM in air-interface designs, focusing on channel prediction and beamforming. These innovative frameworks and applications offer valuable insights and solutions for advancing 6G technology.

Problem

• Deploying large AI models on resource-limited edge networks
• Managing communication and computation for distributed edge LAMs
• Applying edge LAMs to 6G air-interface design challenges

Innovation

• Collaborative fine-tuning and full-parameter training frameworks
• Microservice-assisted inference architecture for edge LAM
• Edge LAM applications in channel prediction and beamforming