TACO: Rethinking Semantic Communications with Task Adaptation and Context Embedding

--Recent advancements in generative artificial intelligence have introduced groundbreaking approaches to innovating next-generation semantic communication, which prioritizes conveying the meaning of a message rather than merely transmitting raw data. A fundamental challenge in semantic communication lies in accurately identifying and extracting the most critical semantic information while adapting to downstream tasks without degrading performance, particularly when the objective at the receiver may evolve over time. T o enable flexible adaptation to multiple tasks at the receiver, this work introduces a novel semantic communication framework, which is capable of jointly capturing task-specific information to enhance downstream task performance and contextual information. Through rigorous experiments on popular image datasets and computer vision tasks, our framework shows promising improvement compared to existing work, including superior performance in downstream tasks, better generalizability, ultra-high bandwidth efficiency, and low reconstruction latency. Next-generation communication systems are expected to support the surge in data-intensive applications with the increasing demand to handle a copious amount of multimodal data generated from intelligent devices, including those from smart sensors, ecosystems of the Internet of Things, mixed reality, and autonomous vehicles [1]. To enable wireless communications with the capacity to satisfy the request from the receiver end with ultra-high bandwidth efficiency in the big data era, semantic communication (SemCOM) has emerged as a transformative paradigm, which shifts data transmission from faithful bitwise recovery of source data to conveying its most critical semantic meaning [2].