dorsal/arxiv
View SchemaLarge Artificial Intelligence Models for Future Wireless Communications
| Authors | Chong Huang, Gaojie Chen, Pei Xiao, Zhu Han, Rahim Tafazolli |
|---|---|
| Categories | |
| ArXiv ID | 2601.06906vv1 |
| URL | https://arxiv.org/abs/2601.06906 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
The anticipated integration of large artificial intelligence (AI) models with wireless communications is estimated to usher a transformative wave in the forthcoming information age. As wireless networks grow in complexity, the traditional methodologies employed for optimization and management face increasingly challenges. Large AI models have extensive parameter spaces and enhanced learning capabilities and can offer innovative solutions to these challenges. They are also capable of learning, adapting and optimizing in real-time. We introduce the potential and challenges of integrating large AI models into wireless communications, highlighting existing AIdriven applications and inherent challenges for future large AI models. In this paper, we propose the architecture of large AI models for future wireless communications, introduce their advantages in data analysis, resource allocation and real-time adaptation, discuss the potential challenges and corresponding solutions of energy, architecture design, privacy, security, ethical and regulatory. In addition, we explore the potential future directions of large AI models in wireless communications, laying the groundwork for forthcoming research in this area.
{
"annotation_id": "7f913349-5940-4f0e-bff1-a064b30a592c",
"date_created": "2026-02-17T05:53:07.891000Z",
"date_modified": "2026-02-17T05:53:07.891000Z",
"file_hash": "34374e75fe33e525345ad7e664248031ba6ee0cb97e2954fdf15f5393e948427",
"private": false,
"record": {
"abstract": "The anticipated integration of large artificial intelligence (AI) models with wireless communications is estimated to usher a transformative wave in the forthcoming information age. As wireless networks grow in complexity, the traditional methodologies employed for optimization and management face increasingly challenges. Large AI models have extensive parameter spaces and enhanced learning capabilities and can offer innovative solutions to these challenges. They are also capable of learning, adapting and optimizing in real-time. We introduce the potential and challenges of integrating large AI models into wireless communications, highlighting existing AIdriven applications and inherent challenges for future large AI models. In this paper, we propose the architecture of large AI models for future wireless communications, introduce their advantages in data analysis, resource allocation and real-time adaptation, discuss the potential challenges and corresponding solutions of energy, architecture design, privacy, security, ethical and regulatory. In addition, we explore the potential future directions of large AI models in wireless communications, laying the groundwork for forthcoming research in this area.",
"arxiv_id": "2601.06906",
"authors": [
"Chong Huang",
"Gaojie Chen",
"Pei Xiao",
"Zhu Han",
"Rahim Tafazolli"
],
"categories": [
"cs.IT",
"eess.SP",
"math.IT"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "Large Artificial Intelligence Models for Future Wireless Communications",
"url": "https://arxiv.org/abs/2601.06906",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "738769ab-d9aa-4cce-8926-067282ac939e",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}