发表机构
IDCoM, University of Edinburgh; KTH Royal Institute of Technology; SnT, University of Luxembourg; Department of Electronic Systems, Aalborg University; School of Computer Science and Informatics, University of Liverpool; Division of Electronics and Electrical Engineering, Dongguk University; School of Engineering and Physical Sciences, Heriot-Watt University; Department of Electrical Engineering, Chalmers University of Technology; Department of Electrical and Computer Engineering, San Diego State University（爱丁堡大学IDCoM研究所; 瑞典皇家理工学院; 卢森堡大学SnT; 奥尔堡大学电子系统系; 利物浦大学计算机科学与信息学院; 东国大学电子与电气工程系; 赫瑞瓦特大学工程与物理科学学院; 查尔姆斯理工大学电气工程系; 圣地亚哥州立大学电气与计算机工程系）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文围绕用于6G智能的联合嵌入预测架构（JEPA）展开。介绍其训练机制等，通过波束管理案例说明能提升标签效率与鲁棒性，还指出在多时间尺度预测等方面存在开放挑战。
AI 中文摘要
第六代（6G）网络正朝着人工智能原生运行发展，学习模块嵌入于无线接入网（RAN）、边缘和核心。这种转变需要从有限标签、异构无线和网络数据、部分观测、非平稳传播以及延迟受限控制回路中学习。联合嵌入预测架构（JEPA）是适用于此场景的有前景的自监督范式，它在潜在空间预测缺失或未来表示。本文给出面向6G智能的JEPA无线导向教程。定义JEPA训练机制，描述如何对信道状态信息（CSI）、波束测量、关键性能指标（KPI）、拓扑图和传感观测进行令牌化和掩码处理，将学习到的编码器定位为用于RAN、O-RAN、边缘和核心功能的预测表示层。通过波束管理案例研究表明，无线感知目标能提高标签效率和鲁棒性。最后概述了多时间尺度预测、动作条件建模、分布式训练、可信度、高效部署、基准测试和标准化等开放挑战。
英文摘要
Sixth-generation (6G) networks are moving toward AI-native operation, where learning modules are embedded across the radio access network (RAN), edge, and core. This transition requires learning from limited labels, heterogeneous wireless and network data, partial observations, non-stationary propagation, and latency-constrained control loops. Joint-embedding predictive architecture (JEPA) is a promising self-supervised paradigm for this setting because it predicts missing or future representations in latent space instead of reconstructing raw measurements or using contrastive negative samples. This article presents a wireless-oriented tutorial on JEPA for 6G intelligence. We define the JEPA training mechanism, describe how CSI, beam measurements, KPIs, topology graphs, and sensing observations can be tokenized and masked, and position the learned encoder as a predictive representation layer for RAN, O-RAN, edge, and core functions, with task-specific heads or controllers producing final decisions. Then we present an illustrative, beam-management case study suggesting that a wireless-aware target, specifically an auxiliary future beam-energy target during self-supervised pretraining, can improve label efficiency and robustness across shifted deployment conditions relative to a supervised source domain. Finally, we outline open challenges in multi-timescale prediction, action-conditioned modeling, distributed training, trustworthiness, efficient deployment, benchmarking, and standardization.
Comments14 pages, 4 figures, 3 tables. Tutorial and review on Joint-Embedding Predictive Architecture (JEPA) for AI-native 6G. Submitted to IEEE Communications Magazine
