面向科研写作辅助的自动化聚焦反馈生成
Automated Focused Feedback Generation for Scientific Writing Assistance
AI总结：
本文提出自动化聚焦反馈生成任务及工具 SWIF$^{2}$T，利用多个 LLM 组件生成具体可操作的论文修改评论，在 300 篇同行评审数据集上验证其优于其他方法，为 AI 反馈整合提供机会。
摘要：
科研写作是一项具有挑战性的任务，对于经常依赖经验丰富同行反馈的 novice 研究人员尤为如此。近期工作主要聚焦于改进表面形式和风格，而非稿件内容。本文提出一项新任务：面向科研写作辅助的自动化聚焦反馈生成。我们提出 SWIF$^{2}$T：一种科研写作聚焦反馈工具。它旨在生成具体、可操作且连贯的评论，以识别科学论文中的弱点并提出修改建议。该方法由 planner、investigator、reviewer 和 controller 四个组件组成，利用多个 Large Language Models (LLMs) 实现这些组件。我们编制了一个包含 300 篇引用科学论文弱点的同行评审数据集，并进行了人工评估。结果表明，与其他方法相比，SWIF$^{2}$T 的反馈在具体性、阅读理解和总体帮助性方面具有优势。分析中还发现，自动生成的评审被判定为优于人工评审的情况，表明将 AI 生成反馈整合到科研写作中的机会。
英文摘要：
Scientific writing is a challenging task, particularly for novice researchers who often rely on feedback from experienced peers. Recent work has primarily focused on improving surface form and style rather than manuscript content. In this paper, we propose a novel task: automated focused feedback generation for scientific writing assistance. We present SWIF$^{2}$T: a Scientific WrIting Focused Feedback Tool. It is designed to generate specific, actionable and coherent comments, which identify weaknesses in a scientific paper and/or propose revisions to it. Our approach consists of four components - planner, investigator, reviewer and controller - leveraging multiple Large Language Models (LLMs) to implement them. We compile a dataset of 300 peer reviews citing weaknesses in scientific papers and conduct human evaluation. The results demonstrate the superiority in specificity, reading comprehension, and overall helpfulness of SWIF$^{2}$T's feedback compared to other approaches. In our analysis, we also identified cases where automatically generated reviews were judged better than human ones, suggesting opportunities for integration of AI-generated feedback in scientific writing.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
