arXiv：cs.CL· Sebastian Joseph, Zichao Wang, Jennifer Healey, Alexa Siu, Junyi Jessy Li, Ani Nenkova·· 4 小时前AI 评分35
GPT 模型生成 AI 论文修改问题：ICLR 与 NeurIPS 投稿对照研究
Generating Edit-Inducing Questions for AI Research Manuscripts
AI 导读
研究对比 GPT 模型与人类审稿人对 ICLR 和 NeurIPS 论文投稿提出的修改问题，GPT 能生成更多引发修改的问题，带来更广泛的内容修改，但其中真正能引发修改的比例远低于人类审稿人。研究确认自动生成问题对作者有帮助，并指出长上下文处理会削弱推理模型产出有用输出的能力。
正文
Abstract:We study the ability of LLMs to generate edit-inducing questions whose answer will improve a paper draft. On a dataset of paired submission and camera-ready papers from ICLR and NeurIPS, we compare the helpfulness of questions from GPT models with or without full paper context to that of human reviewers. GPT produces more edit-inducing questions and its questions are associated with more extensive edits and cover a broader range of edited content compared to questions from reviewers. However, a much smaller percentage of the GPT questions are edit-inducing. Our analyses confirm that automated questions can be beneficial to authors and highlight an example task where proper attending to long context deteriorates reasoning model ability to produce helpful output.
| Comments: | Accepted at the DocInsights Workshop @ EMNLP 2026 |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2609.36617 [cs.CL] |
| (or arXiv:2609.36617v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2609.36617 arXiv-issued DOI via DataCite |
Submission history
From: Sebastian Joseph [view email]
[v1]
Tue, 29 Sep 2026 03:29:57 UTC (272 KB)
[v2]
Wed, 7 Oct 2026 16:09:39 UTC (252 KB)
来源：arXiv：cs.CL · arxiv.org