用隐形字符标记文本，可检测大模型是否偷偷用了受版权保护的内容。
Data Provenance Auditing of Fine-Tuned Large Language Models with a Text-Preserving Technique
- 用看不见的Unicode字符构造'提示-回复'配对作为数字标记。
- 在审计时仅发提示部分，若模型复现回复即表明使用过标记数据。
- 抗误报能力强，适合追踪版权内容在模型训练中的使用情况。
我们提出一种系统，在黑盒访问条件下，通过统计保障检测大语言模型微调中对敏感或受版权保护文本的使用。方法利用隐形Unicode字符构建（'提示', '回复'）配对作为数字标记。审计时仅发送'提示'片段，若模型生成对应'回复'，则表明该文档被使用。为控制误报率，对比保留的反事实标记并应用排名检验，获得可验证的误报率上限。该方法侵入性极低，可扩展至大量数据源，对标准处理流程鲁棒，即使标记数据占微调语料极小比例仍具高检测力。
原文摘要 · Abstract (English)
We propose a system for marking sensitive or copyrighted texts to detect their use in fine-tuning large language models under black-box access with statistical guarantees. Our method builds digital ``marks'' using invisible Unicode characters organized into (``cue'', ``reply'') pairs. During an audit, prompts containing only ``cue'' fragments are issued to trigger regurgitation of the corresponding ``reply'', indicating document usage. To control false positives, we compare against held-out counterfactual marks and apply a ranking test, yielding a verifiable bound on the false positive rate. The approach is minimally invasive, scalable across many sources, robust to standard processing pipelines, and achieves high detection power even when marked data is a small fraction of the fine-tuning corpus.
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。