用提示词检测生成模型是否用了未经授权的数据
CAP: Detecting Unauthorized Data Usage in Generative Models via Prompt Generation
- 设计提示词诱导模型暴露版权内容
- 在四个物联网场景中验证有效
- 适合关注数据合规的AI研发者
为实现准确且无偏的预测，机器学习模型依赖大规模、异构且高质量的数据集。然而，从互联网收集信息可能引发版权和授权方面的伦理与法律问题。随着生成模型的发展，追踪数据来源变得尤为重要，因为它们可能（无意中）复制受版权保护的内容。为此，本文提出版权审计提示生成框架（CAP），用于自动检测机器学习模型是否使用了未经授权的数据。具体而言，我们设计了一种方法，生成能诱导模型揭示版权内容的特定提示词。通过在四个物联网场景中收集的测量数据开展广泛评估，结果表明，该方法在真实和合成数据集上均表现出色。
原文摘要 · Abstract (English)
To achieve accurate and unbiased predictions, Machine Learning (ML) models rely on large, heterogeneous, and high-quality datasets. However, this could raise ethical and legal concerns regarding copyright and authorization aspects, especially when information is gathered from the Internet. With the rise of generative models, being able to track data has become of particular importance, especially since they may (un)intentionally replicate copyrighted contents. Therefore, this work proposes Copyright Audit via Prompts generation (CAP), a framework for automatically testing whether an ML model has been trained with unauthorized data. Specifically, we devise an approach to generate suitable keys inducing the model to reveal copyrighted contents. To prove its effectiveness, we conducted an extensive evaluation campaign on measurements collected in four IoT scenarios. The obtained results showcase the effectiveness of CAP, when used against both realistic and synthetic datasets.
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。