一篇新发表在arXiv上的研究调查了在从临床记录中提取结构化数据时，大语言模型对提示、模型和模式选择的敏感程度。研究人员发现，虽然提示的变化对分类有中等影响，但模型大小的选择显著影响了主要入院标签的重新分配。研究还显示，将提取模式折叠成二进制格式可以解决大多数分歧，这表明缺失与沉默之间的区别是模型解释的关键因素。 AI
影响 强调了在医疗保健等敏感领域仔细配置和验证大语言模型的必要性，以确保可靠的数据提取。
排序理由 学术论文，详细介绍了LLM性能的方法和发现。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2606.05970v1 Announce Type: new Abstract: Large language models are increasingly used for structured extraction from clinical free-text notes, but the sensitivity of their output to upstream configuration choices is less understood than their accuracy on fixed benchmarks. T…
Large language models are increasingly used for structured extraction from clinical free-text notes, but the sensitivity of their output to upstream configuration choices is less understood than their accuracy on fixed benchmarks. This work measures that sensitivity without human…