Adaptive RAN Slicing Control via Reward-Free Self-Finetuning Agents

Researchers propose a novel self-finetuning framework for AI agents that enables continuous learning without handcrafted rewards, demonstrating superior performance in dynamic Radio Access Network slicing tasks. The approach uses bi-perspective reflection to generate autonomous feedback and distill long-term experiences into model parameters, outperforming traditional reinforcement learning methods.

• →New self-finetuning framework allows AI agents to learn continuously through direct environment interaction without explicit reward signals.
• →Bi-perspective reflection mechanism generates autonomous linguistic feedback to construct preference datasets from interaction history.
• →Framework outperforms standard reinforcement learning and existing LLM-based agents in sample efficiency and stability.
• →Successfully applied to dynamic Radio Access Network slicing, a complex multi-objective control problem.
• →Research advances AI-native network infrastructure by enabling self-improving generative agents for continuous control tasks.