Interpretable Language Model for Closed-Loop Type 1 Diabetes Control

Researchers developed LLM-T1D, an interpretable insulin pump controller that combines reinforcement learning with large language models to explain its decisions in plain language. Tested on an FDA-approved simulator, the system achieved 73.5% time in blood sugar range while maintaining safety verification, addressing trust issues in black-box artificial pancreas systems for Type 1 Diabetes.

arXiv:2607.14126v1 Announce Type: new Abstract: Type 1 Diabetes T1D is a chronic, life-threatening autoimmune condition characterized by the complete destruction of insulin-producing pancreatic beta cells. While Artificial Pancreas Systems APS powered by Reinforcement Learning RL have shown promise in automating insulin delivery, their black-box'' nature makes it hard for patients and doctors to trust them fully. This paper presents LLM-T1D, a promising approach that combines the precision of RL with the clear, human-like reasoning of Large Language Models LLMs to create a more transparent and reliable insulin pump controller. By training an expert RL system and distilling its knowledge into fine-tuned LLaMA 3.1 8B and Qwen3 8B models, we developed a controller that not only surpasses the RL system's performance but also explains its decisions in plain, understandable language. Tested on the FDA-approved UVA/Padova T1D simulator, the LLM controllers deliver excellent blood sugar control 73.5% Time in Range while maintaining strict formal safety verification against hallucinations.