A survey of domain generalization in AI-enabled semantic communication: Architecture, challenges and future opportunities

The growing integration of artificial intelligence (AI) into wireless communication systems is driving a shift toward semantic communication, an emerging paradigm that prioritizes the exchange of meaning over raw data. However, semantic communication systems face major challenges when deployed across diverse and unseen domains due to variations in language, context, and channel conditions. This survey provides a comprehensive overview of Domain Generalization (DG) as a key enabler for improving the robustness and adaptability of AI-enabled semantic communication. We explore the types of domain shifts and review the latest DG techniques applicable to semantic communication. Additionally, the paper discusses architectural considerations and real world applications across varied wireless scenarios. Unlike prior works, this survey brings together DG strategies specifically within the context of semantic communication, identifying open challenges and future research directions such as scalable adaptation, resource efficient deployment, and resilience in dynamic environments. It aims to serve as a timely resource for researchers and practitioners working to develop reliable, generalizable communication systems for next generation networks.