Survey Maps Wireless Foundation Models for AI-Native 6G Networks

A new comprehensive survey on arXiv (2608.14694) reviews the emerging field of wireless foundation models (WFMs) for AI-native sixth-generation (6G) networks. The survey, announced as a new paper, addresses the fragmented research landscape by providing a unified taxonomy and framework for designing, learning, and deploying WFMs. Unlike conventional deep learning models trained for individual tasks, WFMs learn generalized representations from large-scale heterogeneous wireless data, enabling efficient adaptation to communication, sensing, localization, and network optimization with minimal task-specific supervision. The authors establish fundamental concepts and introduce a taxonomy that organizes the field, aiming to consolidate current knowledge and guide future research. The paper is available on arXiv, a preprint server, and represents a significant step toward standardizing the development of AI-native 6G technologies.