Nature期刊今日在专栏发表“ChatGPT通过三种方式帮助我进行学术写作”的文章,里面详细介绍了如何使用ChatGPT等生成式人工智能以更好地服务自己的科研工作。今天我们一起学习下~~~
论文润色
语境为王。如果没有语境,就不能指望生成式人工智能(或者任何事物或任何人)对问题做出有意义的回答。当使用聊天机器人完善论文的某个部分以使其更清晰时,首先要概述语境。包括你的论文是关于什么的,你的主要论点是什么?以任何形式记下你的想法——即使是要点也可以。然后,将这些信息呈现给所选择的生成式人工智能。在这篇文章中也给出了示例可供大家参考:
“I’m writing a paper on [topic] for a leading [discipline] academic journal. What I tried to say in the following section is [specific point]. Please rephrase it for clarity, coherence and conciseness, ensuring each paragraph flows into the next. Remove jargon. Use a professional tone.”
AI的第一次回复可能并不完美——这是一个协作和迭代的过程。这意味着我们可能需要改进指示或添加更多信息,就像我们与同事讨论概念时一样。正是这种互动改善了结果。如果有些事情没有完全达到目标,不要犹豫说,“This isn’t quite whatI meant. Let’s adjustthis part.”或者你可以赞扬它的改进:“This is much clearer, butlet’s tweak the ending for a stronger transition to the next section.”
同行评审
生成式人工智能在同行评审过程中可以成为一种有价值的工具。在仔细阅读稿件后,总结出需要评审的要点和领域。然后,使用人工智能帮助组织和表达您的反馈。示例语句:
“Assume you’re an expert and seasoned scholar with 20+ years of academic experience in [field]. On the basis of my summary of a paper in [field], where the main focus is on [generaltopic], provide a detailed review ofthis paper, in the following order: 1) briefly discuss its core content; 2)identify its limitations; and 3) explain the significance of each limitation in order ofimportance. Maintain a concise and professionaltone throughout.”
然而,偶尔它的建议是错误的、牵强的、无关紧要的或完全错误的。这就是为什么审稿的最终责任始终在审稿人身上。审稿人必须能够区分什么是事实,什么不是事实,而没有一个聊天机器人可以可靠地做到这一点。
优化编辑反馈
作为期刊编辑向作者提供建设性的编辑反馈可能具有挑战性,尤其是当编辑每周要审阅多份稿件时。生成式人工智能有助于避免无益的、不具体的反馈,例如,“After careful consideration, we have decided notto proceed with your manuscript”。这篇文章同样给出了指令示例:
“On the basis ofthese notes, draft a letter to the author. Highlightthe manuscript’s key issues and clearly explain why the manuscript, despite its interesting topic, might not provide a substantial enough advancement to merit publication. Avoid jargon. Be direct. Maintain a professionalCareers Machine learning Authorship and respectfultone throughout.”