arXiv 2510.060932025-10-07cs.AI

经典AI与LLMs在健康保险选择中决策者对齐的比较

Classical AI vs. LLMs for Decision-Maker Alignment in Health Insurance Choices

Information Science, Drexel University, Philadelphia, PA 19104 USA；Parallax Advanced Research, 4035 Colonel Glenn Hwy, Beavercreek, OH 45431 USA；Knexus Research, 174 Waterfront Street, Suite 310, National Harbor, Oxon Hill, MD 20745 USA；Information Science \& Computer Science, Drexel University, Philadelphia, PA 19104 USA

Mallika Mainali, Harsha Sureshbabu, Anik Sen, Christopher B. Rauch, Noah D. Reifsnyder, John Meyer, J. T. Turner, Michael W. Floyd, Matthew Molineaux, Rosina O. Weber

备注：: 15 pages, 3 figures. Accepted at the Twelfth Annual Conference on Advances in Cognitive Systems (ACS 2025)
发表信息：: Twelfth Annual Advances in Cognitive Systems Poster Collection (2025), pp. 84-98

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本研究比较经典AI与LLM方法在健康保险决策中对齐不同风险偏好决策者的效果，发现两者对齐程度相当，经典AI在中等风险下略优。

摘要：

随着算法决策者越来越多地应用于高风险领域，AI对齐研究已从关注普遍价值对齐演变为考虑决策者属性的情境特定方法。先前关于决策者对齐（DMA）的工作探索了两种主要策略：（1）结合案例推理、贝叶斯推理和自然决策的经典AI方法，以及（2）利用提示工程的大语言模型（LLM）方法。虽然这两种方法在医疗分诊等有限领域显示出潜力，但它们在新型情境中的泛化能力仍未得到充分探索。在本工作中，我们实现了一个先前的经典AI模型，并开发了一个基于LLM的算法决策者，使用大型推理模型（GPT-5）和非推理模型（GPT-4）在零样本提示框架下进行加权自一致性评估，如近期文献所提议。我们在一个健康保险决策数据集上评估了这两种方法，该数据集针对三个具有不同风险容忍度（0.0、0.5、1.0）的目标决策者进行了标注。在本文报告的实验中，经典AI和基于LLM的模型在基于属性的目标上实现了相当的对齐，其中经典AI在中等风险画像上表现出略好的对齐。数据集和开源实现可在以下网址公开获取：https://github.com/TeX-Base/ClassicalAIvsLLMsforDMAlignment 和 https://github.com/Parallax-Advanced-Research/ITM/tree/feature_insurance。

英文摘要：

As algorithmic decision-makers are increasingly applied to high-stakes domains, AI alignment research has evolved from a focus on universal value alignment to context-specific approaches that account for decision-maker attributes. Prior work on Decision-Maker Alignment (DMA) has explored two primary strategies: (1) classical AI methods integrating case-based reasoning, Bayesian reasoning, and naturalistic decision-making, and (2) large language model (LLM)-based methods leveraging prompt engineering. While both approaches have shown promise in limited domains such as medical triage, their generalizability to novel contexts remains underexplored. In this work, we implement a prior classical AI model and develop an LLM-based algorithmic decision-maker evaluated using a large reasoning model (GPT-5) and a non-reasoning model (GPT-4) with weighted self-consistency under a zero-shot prompting framework, as proposed in recent literature. We evaluate both approaches on a health insurance decision-making dataset annotated for three target decision-makers with varying levels of risk tolerance (0.0, 0.5, 1.0). In the experiments reported herein, classical AI and LLM-based models achieved comparable alignment with attribute-based targets, with classical AI exhibiting slightly better alignment for a moderate risk profile. The dataset and open-source implementation are publicly available at: https://github.com/TeX-Base/ClassicalAIvsLLMsforDMAlignment and https://github.com/Parallax-Advanced-Research/ITM/tree/feature_insurance.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。