这款提示词探讨机关公文写作的语料库构建方法,用户想提供领导讲话、工作报告、调研报告、通知、理论文章等范文,让 ChatGPT 提供指定主题素材并对材料进行校对、润色、续写,同时询问数据清洗、标记、去重、格式转换、标注等环节可用的辅助工具和自动化方案,并咨询在 ChatGPT 中新建「公文写作」对话、大量输入范文素材以提升回答符合度的做法是否可行有效及如何改进,适用于公文写作提效。

中文版提示词

我的需求是提供机关公文的范文,比如领导讲话、工作报告、调研报告、通知、理论文章等,使用场景是写作机关公文时请 ChatGPT 提供指定主题的素材,并对我的材料进行校对、润色、续写等。这样的话,应该怎样处理数据?有没有处理语料库的辅助工具软件?比如数据清洗、标记、去重、格式转换、标注,有的环节是不是可以自动化?另外,我突然有个想法,请帮我分析是否可行:我在 ChatGPT 中新建一个对话《公文写作》,在对话框中输入大量范文内容和素材内容,相当于让 ChatGPT 学习了相关信息,以后需要 ChatGPT 提供素材或文章时,我仍在这个《公文写作》对话中提问,ChatGPT 是不是就可以运用这些范文中的信息,使回答更符合我的需要?这样做是否可行、有效?如果可行,怎样改进才能更好地发挥作用?

英文版提示词

My need is to provide model texts of official government documents, such as leadership speeches, work reports, research reports, notices, and theoretical articles. The use case is: when writing official documents, ask ChatGPT to provide material on a specified topic and to proofread, polish, and continue writing my materials. How should the data be handled in this case? Are there auxiliary tools for processing the corpus, such as data cleaning, tagging, deduplication, format conversion, and annotation? Can some steps be automated? Also, I just had an idea—please help me analyze whether it is feasible: I create a new ChatGPT conversation called "Official Document Writing" and enter a large amount of model texts and material into it, essentially letting ChatGPT learn the relevant information. Later, when I need material or articles from ChatGPT, I still ask questions in this "Official Document Writing" conversation. Would ChatGPT then be able to use the information in these model texts to make its answers better fit my needs? Is this feasible and effective? If so, how can this method be improved to work better?

🛠️ **适用 AI 工具**:ChatGPT、Claude、Kimi、豆包、文心一言、DeepSeek