用大模型和混合相似度，自动从复杂表格中精准提取药代动力学数据。
AutoPK: Leveraging LLMs and a Hybrid Similarity Metric for Advanced Retrieval of Pharmacokinetic Data from Complex Tables and Documents
- 分两阶段：先识别参数变体，再转为标准化键值格式
- 对半衰期和清除率的F1分数达0.92和0.91，优于直接调用大模型
- 让开源模型超越商用系统，适合药物安全与公共健康应用
药代动力学（PK）在人类与兽用药物研发及监管决策中至关重要，直接影响用药安全与疗效评估。然而，PK数据常嵌入结构多样、术语不一的复杂表格中，自动化提取与标准化面临挑战。本文提出AutoPK，一种两阶段框架，可准确且可扩展地从复杂科学表格中提取PK数据。第一阶段利用大语言模型（LLMs）、混合相似度度量及基于LLM的验证，识别并提取PK参数变体；第二阶段筛选相关行，将表格转换为键值文本格式，并通过LLM重建标准化表格。在包含605张真实世界PK表格（含图注和脚注）的数据集上评估，AutoPK显著优于直接使用LLM的基线。例如，采用LLaMA 3.1-70B时，半衰期与清除率的F1得分分别为0.92和0.91，较直接使用提升0.10和0.21。较小模型如Gemma 3-27B与Phi 3-12B在使用AutoPK后，F1得分提升2至7倍，幻觉率由60%-95%降至8%-14%。值得注意的是，Gemma 3-27B在多个参数上表现超过GPT-4o Mini。AutoPK实现高置信度、可扩展的PK数据提取，适用于兽药药理、药物安全监测与公共卫生决策，能有效应对表格异构性与术语差异，并在关键PK参数上展现泛化能力。代码与数据：https://github.com/hosseinsholehrasa/AutoPK
原文摘要 · Abstract (English)
Pharmacokinetics (PK) plays a critical role in drug development and regulatory decision-making for human and veterinary medicine, directly affecting public health through drug safety and efficacy assessments. However, PK data are often embedded in complex, heterogeneous tables with variable structures and inconsistent terminologies, posing significant challenges for automated PK data retrieval and standardization. AutoPK, a novel two-stage framework for accurate and scalable extraction of PK data from complex scientific tables. In the first stage, AutoPK identifies and extracts PK parameter variants using large language models (LLMs), a hybrid similarity metric, and LLM-based validation. The second stage filters relevant rows, converts the table into a key-value text format, and uses an LLM to reconstruct a standardized table. Evaluated on a real-world dataset of 605 PK tables, including captions and footnotes, AutoPK shows significant improvements in precision and recall over direct LLM baselines. For instance, AutoPK with LLaMA 3.1-70B achieved an F1-score of 0.92 on half-life and 0.91 on clearance parameters, outperforming direct use of LLaMA 3.1-70B by margins of 0.10 and 0.21, respectively. Smaller models such as Gemma 3-27B and Phi 3-12B with AutoPK achieved 2-7 fold F1 gains over their direct use, with Gemma's hallucination rates reduced from 60-95% down to 8-14%. Notably, AutoPK enabled open-source models like Gemma 3-27B to outperform commercial systems such as GPT-4o Mini on several PK parameters. AutoPK enables scalable and high-confidence PK data extraction, making it well-suited for critical applications in veterinary pharmacology, drug safety monitoring, and public health decision-making, while addressing heterogeneous table structures and terminology and demonstrating generalizability across key PK parameters. Code and data: https://github.com/hosseinsholehrasa/AutoPK
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。