Progressive Semantic Communication for Efficient Edge-Cloud Vision-Language Models

TL;DR

This paper introduces a progressive semantic communication framework for edge-cloud vision-language models, enabling adaptive, bandwidth-efficient inference with maintained semantic fidelity on resource-constrained devices.

Contribution

It proposes a Meta AutoEncoder-based adaptive compression scheme that allows flexible, plug-and-play deployment of VLMs without fine-tuning, optimizing communication under bandwidth constraints.

Findings

Significant latency reduction at 1 Mbps uplink compared to baseline methods.

High semantic consistency maintained under high compression levels.

Effective deployment on embedded platforms with end-to-end system demonstrated.

Abstract

Deploying Vision-Language Models (VLMs) on edge devices remains challenging due to their substantial computational and memory demands, which exceed the capabilities of resource-constrained embedded platforms. Conversely, fully offloading inference to the cloud is often impractical in bandwidth-limited environments, where transmitting raw visual data introduces substantial latency overhead. While recent edge-cloud collaborative architectures attempt to partition VLM workloads across devices, they typically rely on transmitting fixed-size representations, lacking adaptability to dynamic network conditions and failing to fully exploit semantic redundancy. In this paper, we propose a progressive semantic communication framework for edge-cloud VLM inference, using a Meta AutoEncoder that compresses visual tokens into adaptive, progressively refinable representations, enabling plug-and-play deployment with off-the-shelf VLMs without additional fine-tuning. This design allows flexible transmission at different information levels, providing a controllable trade-off between communication cost and semantic fidelity. We implement a full end-to-end edge-cloud system comprising an embedded NXP i.MX95 platform and a GPU server, communicating over bandwidth-constrained networks. Experimental results show that, at 1 Mbps uplink, the proposed progressive scheme significantly reduces network latency compared to full-edge and full-cloud solutions, while maintaining high semantic consistency even under high compression. The implementation code will be released upon publication at https://github.com/open-ep/ProSemComVLM.