The growing complexity of next-generation networks makes intelligent and adaptive management essential forOpen Radio Access Network (O-RAN)-based network slicing. Traditional Reinforcement Learning (RL) solutions in xAppsautomate resource allocation but struggle to incorporate highlevel operator intents into their optimization process. To addressthis limitation, we propose a Large Language Model (LLM)-guided RL framework that bridges the gap between human intent and autonomous network control. In this approach, the LLM-based agent in the rApp interprets operator promptsand contextual network information to dynamically adjust the reward function of the RL agent operating in the xApp, ensuringthat resource allocation decisions across slices reflect strategic objectives. The simulation results show that the proposed LLMguided RL approach achieves higher cumulative reward while maintaining closer alignment with the operator objectives,creating the basis for more intelligent and dynamic O-RAN management.
Chiarani et al. (Mon,) studied this question.