Introduction
The unprecedented growth in user data and the continuous advancement of generative artificial intelligence (GAI) models have led to ground-breaking applications such as Google Bard and ChatGPT. As users increasingly benefit from these applications, their attention is concurrently shifting to the principles of GAI [1], which powers these applications. Unlike traditional AI (TAI) models that prioritize sample analysis, training, and classification, GAI specializes in understanding and modeling the distribution of complex data. By leveraging statistical properties of the training data, GAI can generate data similar to the training data [2]. For example, the ControINet [3] can generate images with outstanding quality, in terms of resolution and naturalness, demonstrating great efficiency. In the context of the rapid evolving of wireless network services, GAI is poised to meet the various and ever-changing content generation needs of users.