The field of wireless networks is witnessing a significant shift towards AI-driven solutions, with a focus on leveraging large language models (LLMs) and reinforcement learning (RL) to improve network management and resource allocation. Recent developments have highlighted the potential of Transformer-empowered architectures in optimizing service function chain partitioning and enabling decentralized generative AI. The integration of LLMs with RL is also being explored to enhance network intelligence and adaptability. Furthermore, the deployment of large AI models at the edge of wireless networks is being investigated to support real-time intelligent services in 6G networks. Noteworthy papers include the introduction of the LLM-hRIC framework for hierarchical RAN intelligent control, and the proposal of a Transformer-empowered actor-critic framework for sequence-aware service function chain partitioning. Additionally, a comprehensive survey of the Mixture of Experts framework in wireless networks has been conducted, highlighting its potential in optimizing resource efficiency and improving scalability.