用幻觉检测引导模型自修正，提升临床摘要真实性
Hallucination Detection-Guided Preference Optimization for Clinical Summarization
- 用幻觉检测器指导摘要迭代修改，自动发现并纠正错误
- 对Llama-3.1-8B模型可减少48%幻觉，且保持语义流畅性
- 适合医疗AI研发者，解决临床文本生成中的可信度难题
大语言模型在摘要任务中表现良好，但常产生无依据或错误陈述，限制其在医疗领域的可靠性。本文提出霍拉西翁检测引导的自修正方法（HDSR），在推理阶段利用幻觉检测器指导摘要逐步修正。在此基础上，进一步提出HDSR-PL，将检测引导的修正过程转化为偏好对用于模型微调。大量实验表明，该方法显著降低Llama与Gemma模型在真实临床笔记（MIMIC-IV-Note v2.2）摘要中的幻觉率。例如，HDSR使Llama-3.1-8B-Instruct幻觉减少24%，而HDSR-PL实现48%的降幅。重要的是，人类专家与LLM-Jury评估均显示，两种方法在保持摘要流畅性、连贯性和相关性方面表现优异。结果表明，基于检测反馈的修正与偏好学习可提供自动化提升临床摘要事实准确性的方案。
原文摘要 · Abstract (English)
Large language models (LLMs) have shown promise on summarization tasks, but they often produce hallucinations, which are unsupported or incorrect statements that limit their reliability in specialized healthcare applications. We introduce Hallucination Detection Guided Self-Refinement (HDSR), an inference-time method that leverages hallucination detectors to guide iterative summary revisions toward factual corrections. Building on this, we propose HDSR for Preference Learning (HDSR-PL), which converts detector-guided refinement trajectories into preference pairs for model finetuning. Extensive experiments show that our methods substantially reduce hallucinations for Llama and Gemma models in summarizing real-world clinical notes from MIMIC-IV-Note v2.2. For example, HDSR reduces 24% and HDSR-PL reduces 48% hallucinations in Llama-3.1-8B-Instruct. Importantly, both methods preserve summary fluency, coherence, and relevance according to human expert and LLM-Jury evaluations. Together, these results demonstrate that detection-informed refinement and preference learning offer an automated solution for improving factual faithfulness in clinical summarization.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。