arXiv cs.AI· Junchi Liao, Xiao Huang, Yucheng Liu, Jiawen Deng, Fuji Ren·· 9 天前AI 评分66
研究：GPT 模型改写论文时出现防御性写作，GPT-6-astra 会直接撤回作者结论
Defensive Writing in GPT Models
AI 导读
一项研究提出防御性写作概念，指 GPT 模型在材料不支持的情况下收窄甚至撤回作者结论，并测试了两种解释。研究让 GPT 各版本及其他厂商模型改写 ChatGPT 之前论文的段落，或依据列出方法与结果的事实表写作，发现防御性写作随 GPT 版本增强，GPT-6-astra 会直接撤回作者结论，且依据事实表写作时添加的无依据限定最多。
正文 · 原文
Abstract:Researchers increasingly use ChatGPT to revise their papers, and recent GPT versions often narrow or even retract the authors' claims. We call such changes defensive writing when the given material does not support them, and we test two explanations: the model corrects the authors' overclaiming, or it writes for an anticipated reviewer. We ask GPT versions and models from other developers to rewrite paragraphs from papers written before ChatGPT, or to write from an evidence sheet that lists a paper's method and results. Defensive writing grows with GPT version. GPT-6-astra retracts the authors' claims outright, and when it writes from the evidence sheet, it still adds the most ungrounded qualifications. The results favor the anticipated-review explanation, and correcting overclaiming explains only a small part. When the models are only asked to polish, defense stays near the level of the originals; mentioning review raises it, and one round of self-review raises it further. At the same time, fewer than one in ten of the claims GPT-6-astra retracts are overstated. AI reviewers score defensive rewrites higher, while human readers find them harder to read and the authors less certain. Combining AI writing with AI review may amplify this style.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.11355 [cs.AI] |
| (or arXiv:2610.11355v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.11355 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Junchi Liao [view email]
[v1]
Thu, 8 Oct 2026 06:47:48 UTC (229 KB)
来源：arXiv cs.AI · arxiv.org