CE-LSLM: Efficient Large-Small Language Model Inference and Communication via Cloud-Edge Collaboration

TL;DR

This paper introduces a cloud-edge collaborative inference framework for large language models in 6G networks, enhancing efficiency, privacy, and responsiveness by sharing semantic states and optimizing communication and computation.

Contribution

It proposes a novel architecture integrating cloud LLMs with edge small models, featuring key-value cache reuse, cross-node scheduling, and model alignment strategies for efficient edge inference.

Findings

Reduces edge computational and storage overhead.

Improves inference latency and system stability.

Enhances scalability and responsiveness in 6G scenarios.

Abstract

Emerging intelligent service scenarios in 6G communication impose stringent requirements for low latency, high reliability, and privacy preservation. Generative large language models (LLMs) are gradually becoming key enablers for the integration of semantic communication and computation. However, due to the limited computational resources of edge devices and the increasing complexity of heterogeneous terminal access, existing centralized inference approaches fail to meet the dual demands of response efficiency and data privacy in edge-side inference tasks. To address these challenges, this paper proposes a novel collaborative inference architecture that integrates cloud-based LLMs with edge-deployed small language models (SLMs), enabling dynamic scheduling and sharing of semantic-level intermediate states, and establishing a unified computation-communication paradigm tailored for 6G networks. Specifically, a key-value (KV) cache reuse mechanism is introduced to enhance the semantic understanding of edge models through contextual guidance from the cloud, while significantly reducing edge-side computational and storage overhead. Furthermore, a cross-node parallel scheduling mechanism is proposed to achieve asynchronous coordination between model state loading and decoding computation, thereby improving edge responsiveness. In addition, we investigate layer alignment and representation compression strategies between heterogeneous models to alleviate the communication burden on the edge. Experimental results demonstrate that the proposed architecture exhibits superior adaptability and scalability in terms of inference latency, system stability, and concurrent processing capacity.