用真人文本指导改写AI文本，让机器生成更像人写的。
On the Indistinguishability of Human v/s AI Generated Text
- 用真人样本指导反复改写AI文本，使其分布趋近人类
- 在有限样本下仍能保证改写效果，收敛速度可计算
- 适合关注文本检测与防伪的开发者和研究者
大型语言模型的快速进步使区分AI生成文本与人类写作成为紧迫问题。这一挑战因旨在让机器生成文本更“像人”的改写工具而加剧。本文研究如何利用人类写作样本来策略性地改写机器生成内容，使其分布趋近人类分布。在相同提示下同时拥有真人和机器响应的多样本设定中，我们证明在简单混合与稳定性条件下，重复改写能使机器分布向经验上的人类分布收敛。结果给出了明确的收敛速率，将分析扩展至有限样本情形，并刻画了所需真人样本数与改写轮次随目标误差的变化关系。
原文摘要 · Abstract (English)
The rapid improvement of LLMs has made distinguishing AI-generated text from human writing a pressing problem. This challenge is further amplified by paraphrasing tools designed to make machine-generated text appear more "human". We study how access to human writing samples can be used to strategically paraphrase machine-generated responses toward the human distribution. Under a multi-sample setting with human and machine responses to the same prompts, we show that repeated paraphrasing moves the machine distribution toward the empirical human distribution under simple mixing and stability conditions. Our results derive an explicit convergence rate, extend the analysis to a finite-sample setting, and characterize how the required number of human samples and paraphrasing rounds scale with the desired error.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。