Abstract:Large language models (LLMs) and foundation models have been recently touted as a game-changer for 6G systems. However, recent efforts on LLMs for wireless networks are limited to a direct application of existing language models that were designed for natural language processing (NLP) applications. To address this challenge and create wireless-centric foundation models, this paper presents a comprehensive vision on how to design universal foundation models that are tailored towards the deployment of artificial intelligence (AI)-native networks. Diverging from NLP-based foundation models, the proposed framework promotes the design of large multi-modal models (LMMs) fostered by three key capabilities: 1) processing of multi-modal sensing data, 2) grounding of physical symbol representations in real-world wireless systems using causal reasoning and retrieval-augmented generation (RAG), and 3) enabling instructibility from the wireless environment feedback to facilitate dynamic network adaptation thanks to logical and mathematical reasoning facilitated by neuro-symbolic AI. In essence, these properties enable the proposed LMM framework to build universal capabilities that cater to various cross-layer networking tasks and alignment of intents across different domains. Preliminary results from experimental evaluation demonstrate the efficacy of grounding using RAG in LMMs, and showcase the alignment of LMMs with wireless system designs. Furthermore, the enhanced rationale exhibited in the responses to mathematical questions by LMMs, compared to vanilla LLMs, demonstrates the logical and mathematical reasoning capabilities inherent in LMMs. Building on those results, we present a sequel of open questions and challenges for LMMs. We then conclude with a set of recommendations that ignite the path towards LMM-empowered AI-native systems.

Large Language Model Enabled Multi-Task Physical Layer Network

M-Net: A Lightweight Network Based on Multilayer Perceptron for Massive MIMO CSI Feedback

Large Language Model (LLM)-enabled In-context Learning for Wireless Network Optimization: A Case Study of Power Control

Large Multi-Modal Models (LMMs) as Universal Foundation Models for AI-Native Wireless Systems

Large Language Model (LLM) for Telecommunications: A Comprehensive Survey on Principles, Key Techniques, and Opportunities

Integrating Pre-Trained Language Model with Physical Layer Communications

WirelessLLM: Empowering Large Language Models Towards Wireless Intelligence

Large Language Models (LLMs) Assisted Wireless Network Deployment in Urban Settings

When Large Language Models Meet Optical Networks: Paving the Way for Automation

Large Language Models for Networking: Applications, Enabling Techniques, and Challenges

LLM-Slice: Dedicated Wireless Network Slicing for Large Language Models

Joint Deep Reinforcement Learning and Unfolding: Beam Selection and Precoding for Mmwave Multiuser MIMO with Lens Arrays

Personalized Wireless Federated Learning for Large Language Models

The Power of Large Language Models for Wireless Communication System Development: A Case Study on FPGA Platforms

Large Language Model Enabled Semantic Communication Systems

6G comprehensive intelligence: network operations and optimization based on Large Language Models

LLM Agents as 6G Orchestrator: A Paradigm for Task-Oriented Physical-Layer Automation

LLM4CP: Adapting Large Language Models for Channel Prediction

R-MTLLMF: Resilient Multi-Task Large Language Model Fusion at the Wireless Edge

Synergistic Interplay of Large Language Model and Digital Twin for Autonomous Optical Networks: Field Demonstrations