LLMs annotate bioassay metadata for data readiness

1. 12:00
   arXiv · Databases

   LLMs 能否可靠地标注生物实验元数据以提升数据就绪度？

   研究评估了 7 个开源与闭源大语言模型对 PubChem 约 200 万条生物实验的元数据标注能力，发现实验格式缺失率 36%、BioAssay 类型缺失率 89%、BAO 映射术语缺失率 >99.9%。模型在生化与细胞实验格式上的召回率至少达 0.96，多数分歧可追溯至银标来源不一致而非模型错误，且 LLM 证据能促使专家修正自身标签。

Current heat 9·Comparable peak 10（Oct 2）·Comparable change over 24 hours –

The trend compares only the same participants observed continuously; its range may be smaller than the current heat count. Move or click on the chart to inspect hourly heat; use the left and right arrow keys to switch.