科学知识主要存储于静态论文中。一个名为 Paper2Agent 的自动化框架现可将每篇论文转化为主动的人工智能智能体——即虚拟通讯作者，该智能体能够回答问题、将论文方法应用于新数据，并与其他论文智能体协作。这使研究更易于复现、复用和扩展。通过自然语言提示，虚拟通讯作者可以解释并复现论文的分析过程。

Scientific knowledge is mostly stored in static papers. An automated framework called Paper2Agent can now transform each paper into an active artificial intelligence agent — a virtual corresponding author that answers questions, applies the paper’s methods to new data, and collaborates with other paper agents. This makes research easier to reproduce, reuse and extend. A virtual corresponding author prompted in natural language can explain and reproduce a paper’s analyses.