AI 中文总结
研究针对1型糖尿病控制中人工胰腺系统“黑箱”问题，提出LLM-T1D方法，结合强化学习与大语言模型，训练专家RL系统并提炼知识到模型，开发出可解释且性能优的胰岛素泵控制器，在模拟器测试中血糖控制良好且能防幻觉。
AI 中文摘要
1型糖尿病（T1D）是一种慢性、危及生命的自身免疫性疾病，其特征是产生胰岛素的胰腺β细胞完全被破坏。虽然由强化学习（RL）驱动的人工胰腺系统（APS）在自动化胰岛素输送方面显示出前景，但其“黑箱”性质使患者和医生难以完全信任它们。本文提出了LLM-T1D，一种将RL的精确性与大语言模型（LLMs）清晰、类人的推理相结合的有前途的方法，以创建一个更透明、可靠的胰岛素泵控制器。通过训练一个专家RL系统并将其知识提炼到微调后的LLaMA 3.1 8B和Qwen3 8B模型中，我们开发了一个不仅超越RL系统性能，还能用通俗易懂的语言解释其决策的控制器。在FDA批准的UVA/Padova T1D模拟器上进行测试，LLM控制器在保持对幻觉的严格形式安全验证的同时，实现了出色的血糖控制（73.5%的时间在范围内）。
英文摘要
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.
CommentsAccepted at the 2026 IEEE 22nd International Conference on Automation Science and Engineering conference (IEEE CASE 2026)
