小模型生成临床数据结构化输出时，JSON最易解析，需注意文档长度和类型影响。
Evaluating Structured Output Robustness of Small Language Models for Open Attribute-Value Extraction from Clinical Notes
- 对比三种格式，JSON解析成功率最高
- 长文档和特定病历类型导致解析率下降
- 提示词优化可提升结构鲁棒性，适合医疗场景部署
我们对小型语言模型在临床笔记中进行开放属性-值提取时生成的结构化输出的可解析性进行了比较分析。评估了三种常用序列化格式：JSON、YAML 和 XML，发现 JSON 始终具有最高的可解析性。结构鲁棒性随针对性提示词和模型规模增大而提升，但在长文档和某些病历类型下会下降。错误分析揭示了重复出现的格式相关失败模式。这些发现为在隐私敏感的临床环境中选择序列化格式和设计提示词提供了实用指导。
原文摘要 · Abstract (English)
We present a comparative analysis of the parseability of structured outputs generated by small language models for open attribute-value extraction from clinical notes. We evaluate three widely used serialization formats: JSON, YAML, and XML, and find that JSON consistently yields the highest parseability. Structural robustness improves with targeted prompting and larger models, but declines for longer documents and certain note types. Our error analysis identifies recurring format-specific failure patterns. These findings offer practical guidance for selecting serialization formats and designing prompts when deploying language models in privacy-sensitive clinical settings.
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。