Skip to content
Book Open access

TACToC: Enabling Task-Adaptive Communication System with a Task-Describable Encoder

Oct 2026 · Proceedings of the 32nd Annual International Conference on Mobile Computing and Networking · pp. 282-296 · 0 citations · 16 references

Abstract

Task-oriented communications (ToC), also known as semantic communications, offers a promising paradigm for offloading deep neural network (DNN) inference from edge devices to more powerful remote servers. By transmitting only task-specific information, ToC significantly reduces the communication overhead inherent in conventional offloading systems. However, existing ToC approaches exhibit limited task adaptivity, requiring retraining or separate encoders for each task—hindering their practicality in real-world multi-task settings. In this work, we present TACToC, the first ToC system to achieve task adaptivity with a single encoder, enabled by a novel task-describable encoder structure. TACToC first maps input data into a semantically structured feature space using a self-supervised vision transformer (SSL-ViT), providing a foundation for flexible task-specific feature selection. The encoder then dynamically selects and transmits only important features for the target task, without per-task optimization. This is realized through two key mechanisms: (i) relevant feature estimation, which quantifies each feature's contribution to task inference, and (ii) prompt-based task embedding, which injects task-specific information to guide selective transmission for each task. Extensive experiments demonstrate that TACToC reduces communication cost by up to 1.5-14.5× within the range that allows for a 3%p degradation while supporting diverse vision tasks with seamless task integration and transition. This work marks a significant step toward practical and unified ToC systems for dynamic and multi-task AI services.

Read PDF

Similar papers

Preprint Oct 2026

More Interactions Make Communications More Efficient: A Directed Information Bottleneck Approach to Task-Oriented Communication

Experiments demonstrate that VDIB-JSCC achieves higher classification accuracy with less transmitted information than state-of-the-art semantic communication methods and show that leveraging historical knowledge enables task completion with less information, improving communication efficiency.

Zhirun Zheng, Yi-Qin Deng, Phone Lin et al. · 0 citations
Review Oct 2026

Semantic Communications: Principles, Architectures and Applications

Conventional communication systems are designed to transmit data reliably, focusing on metrics such as bit error rate, throughput, and channel capacity. However, these traditional measures do not evaluate whether the received information is actually useful for a specific downstream task. To address this limitation, sem...

Sasa Maric · 0 citations
Open access 2026

DRASST-MTL: Attention Guided Multi-Task Network for Joint Modulation Recognition and Symbol Rate Estimation

In modern wireless communication systems, especially cognitive radio and non-cooperative communication, modulation recognition and symbol rate estimation are critical for accurate demodulation and decoding since modulation recognition is a prerequisite for subsequent signal demodulation, while symbol rate provides esse...

Rui-Feng Duan, Xiang-Yang Liu, Feng Yang et al. · 0 citations
Preprint Nov 2026

UniCompress: From Multi-LoRA Experts to One Compact Model for Edge Intelligence

Embodied and edge systems increasingly rely on vision-language models (VLMs) to support heterogeneous skills, yet deploying a separate task adapter for each skill while retaining a high-precision backbone creates substantial memory and management costs. We present UniCompress, a deployment-oriented framework that conve...

Ming-Kang Liu, P. Jiao, Wei-Ming Meng et al. · 0 citations
Preprint Nov 2026

Over-the-Air Computation for Imperfectly Synchronized Multi-View Semantic Communication

Compared to classical task-agnostic communication systems, which focus on reliable transmission of a message, semantic communication (SemCom) focuses on transmitting the task-relevant meaning behind the message. In this paper, we design a SemCom system for a distributed multi-view scenario, where multiple semantic devi...

Maximilian H. V. Tillmann, J. Sundermann, Ahmad Halimi Razlighi et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.