arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2511.13293cs.AI
收藏

经验锚定:借助分层智能体检索增强的生成式医疗预测

Grounded by Experience: Generative Healthcare Prediction Augmented with Hierarchical Agentic Retrieval

Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology(电子与计算机工程系,香港科技大学);College of Management and Economics, Laboratory of Computation and Analytics of Complex Management Systems (CACMS), Tianjin University(管理学院与经济学学院,复杂管理系统的计算与分析实验室,天津大学);Department of Computer Science and Engineering, The Hong Kong University of Science and Technology(计算机科学与工程系,香港科技大学)

Chuang Zhao, Hui Tang, Hongke Zhao, Xiaofang Zhou, Xiaomeng Li

AI总结:

针对现有RAG框架用于医疗预测时的检索触发判断、检索生成协同难题,提出GHAR分层智能体RAG框架,通过双智能体架构与马尔可夫决策过程统一优化,在三个基准数据集上取得优于SOTA基线的效果。

摘要:

精准的医疗预测对于改善患者预后、降低运营成本至关重要。随着推理能力不断提升,大语言模型(LLMs)凭借其丰富的参数化知识,为优化医疗预测效果提供了极具潜力的路径。然而,受内置知识的可靠性与覆盖范围限制,LLMs容易出现事实性错误。尽管GraphRAG及其变体等检索增强生成(RAG)框架已被提出,可通过引入外部知识缓解上述问题,但这类框架在医疗场景中面临两大核心挑战:一是判断激活检索机制的临床必要性,二是实现检索器与生成器的协同,以生成适配上下文的检索结果。为应对这些挑战,我们提出GHAR——生成式分层智能体RAG(generative hierarchical agentic RAG)框架,可同时解决医疗场景中何时触发检索、如何优化子模块间协作的问题。具体而言,针对第一个挑战,我们设计了包含顶层智能体(Agent-Top)与底层智能体(Agent-Low)的双智能体架构:Agent-Top扮演主治医师角色,迭代决策是依赖参数化知识还是启动检索;Agent-Low扮演咨询服务角色,在检索触发后汇总所有与任务相关的知识。针对第二个挑战,我们创新性地将两个智能体的优化统一在形式化马尔可夫决策过程框架下,设计多样化奖励机制,在保留二者角色差异的同时,对齐其实现精准预测的共同目标。我们在覆盖三类主流任务的三个基准数据集上开展了大量实验,结果表明GHAR性能优于当前最优基线模型,凸显了分层智能体RAG在推进医疗系统发展方面的潜力。

英文摘要:

Accurate healthcare prediction is critical for improving patient outcomes and reducing operational costs. Bolstered by growing reasoning capabilities, large language models (LLMs) offer a promising path to enhance healthcare predictions by drawing on their rich parametric knowledge. However, LLMs are prone to factual inaccuracies due to limitations in the reliability and coverage of their embedded knowledge. While retrieval-augmented generation (RAG) frameworks, such as GraphRAG and its variants, have been proposed to mitigate these issues by incorporating external knowledge, they face two key challenges in the healthcare scenario: (1) identifying the clinical necessity to activate the retrieval mechanism, and (2) achieving synergy between the retriever and the generator to craft contextually appropriate retrievals. To address these challenges, we propose GHAR, a \underline{g}enerative \underline{h}ierarchical \underline{a}gentic \underline{R}AG framework that simultaneously resolves when to retrieve and how to optimize the collaboration between submodules in healthcare. Specifically, for the first challenge, we design a dual-agent architecture comprising Agent-Top and Agent-Low. Agent-Top acts as the primary physician, iteratively deciding whether to rely on parametric knowledge or to initiate retrieval, while Agent-Low acts as the consulting service, summarising all task-relevant knowledge once retrieval was triggered. To tackle the second challenge, we innovatively unify the optimization of both agents within a formal Markov Decision Process, designing diverse rewards to align their shared goal of accurate prediction while preserving their distinct roles. Extensive experiments on three benchmark datasets across three popular tasks demonstrate our superiority over state-of-the-art baselines, highlighting the potential of hierarchical agentic RAG in advancing healthcare systems.

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。

↑