A statistical framework of watermarks for large language models: Pivot, detection efficiency and optimal rules
大语言模型水印的统计框架：枢轴量、检测效率与最优规则AI 解读 ↓
← 返回论文列表与筛选
阅读说明
本页整理公开书目信息，不生成学术质量评分。AI 解读根据题名及所注明的摘要或原文片段生成，未完成全文核验，可能有误；具体假设、证明与结论请核对原文。摘要保留来源文本，未经模型改写。
DOI 首次登记不等于接收或发表日期。部分更正、评论等条目没有独立摘要；缺少可靠材料时保留待补状态。
引用量采用 OpenAlex 单条关联记录，不合计多个记录；关联记录可能包含正式发表版本，不等于此版本单独引用。缺失不视为零，引用指标不代表论文质量。
关注作者按书目中的完整姓名匹配，同名和姓名变体需人工辨别。相关论文先按共享细分方向和具体方法排列，再参考通用方法与发表日期。
分类依据与纠错
AI 根据所提供的论文材料归类：摘要研究语言模型文本水印与通用假设检验，没有所列经济计量细分方向的直接依据。
反馈分类问题正在核对关联版本…
代码、数据与补充材料
正在核对对应版本的资源…
大语言模型水印的统计框架：枢轴量、检测效率与最优规则
- 研究问题
- 如何在控制将人类文本误判为模型文本的概率时，评价语言模型水印的检测效率，并设计具有较高功效的规则？
- 方法与思路
- 作者利用文本与密钥构造枢轴统计量，推导渐近假阴性率表达，再通过极小极大优化确定框架内的最优检测规则。
- 材料陈述的贡献
- 摘要称框架适用于两种代表性水印，推导相应最优规则；数值实验显示其表现具有竞争力，部分情形下功效更高。
- 阅读关注点
- 应核对枢轴性质和错误率控制的条件、极小极大优化中的对手范围，以及文本长度和水印机制如何影响检测功效。
- 阅读边界
- 仅据版本未注明的索引摘要；最优性限于给定框架，摘要未说明改写攻击或分布变化下的保证，不能泛化数值优势。
原文摘要
Since ChatGPT was introduced in November 2022, embedding (nearly) unnoticeable statistical signals into text generated by large language models (LLMs), also known as watermarking, has been used as a principled approach to provable detection of LLM-generated text from its human-written counterpart. In this paper, we introduce a general and flexible framework for reasoning about the statistical efficiency of watermarks and designing powerful detection rules. Inspired by the hypothesis testing formulation of watermark detection, our framework starts by selecting a pivotal statistic of the text and a secret key-provided by the LLM to the verifier-to control the false positive rate (the error of mistakenly detecting human-written text as LLM-generated). Next, this framework allows one to evaluate the power of watermark detection rules by obtaining a closed-form expression of the asymptotic false negative rate (the error of incorrectly classifying LLM-generated text as human-written). Our framework further reduces the problem of determining the optimal detection rule to solving a minimax optimization program. We apply this framework to two representative watermarks-one of which has been internally implemented at OpenAI-and obtain several findings that can be instrumental in guiding the practice of implementing watermarks. In particular, we derive optimal detection rules for these watermarks under our framework. These theoretically derived detection rules are demonstrated to be competitive and sometimes enjoy a higher power than existing detection approaches through numerical experiments.