DOI: 10.1145/3543873.3587321

Copyright Protection and Accountability of Generative AI:Attack, Watermarking and Attribution

Haonan ZhongJiamin ChangZiyue YangTingmin WuPathum Chamikara Mahawaga ArachchigeChehara PathmabanduMinhui Xue

Haonan ZhongJiamin ChangZiyue Yang ...+3 Minhui Xue

Mar 2023

摘要

Generative AI (e.g., Generative Adversarial Networks - GANs) has becomeincreasingly popular in recent years. However, Generative AI introducessignificant concerns regarding the protection of Intellectual Property Rights(IPR) (resp. model accountability) pertaining to images (resp. toxic images)and models (resp. poisoned models) generated. In this paper, we propose anevaluation framework to provide a comprehensive overview of the current stateof the copyright protection measures for GANs, evaluate their performanceacross a diverse range of GAN architectures, and identify the factors thataffect their performance and future research directions. Our findings indicatethat the current IPR protection methods for input images, model watermarking,and attribution networks are largely satisfactory for a wide range of GANs. Wehighlight that further attention must be directed towards protecting trainingsets, as the current approaches fail to provide robust IPR protection andprovenance tracing on training sets.

展开全部

图表提取

论文十问

参考文献

被引用

提问与回答

暂无人提供速读十问回答

被引用

笔记

问答

领域