WASA: WAtermark-based Source Attribution for Large Language Model-Generated Data

Xinyang Lu, Jingtan Wang, Zitong Zhao, Zhongxiang Dai, Chuan-Sheng Foo, See-Kiong Ng, Bryan Kian Hsiang Low

Primary Area: societal considerations including fairness, safety, privacy

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Keywords: Large Language Model, Source Attribution, Data Provenance

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

Abstract: The impressive performances of large language models (LLMs) and their immense potential for commercialization have given rise to serious concerns over the intellectual property (IP) of their training data. In particular, the synthetic texts generated by LLMs may infringe the IP of the data being used to train the LLMs. To this end, it is imperative to be able to (a) identify the data provider who contributed to the generation of a synthetic text by an LLM (source attribution) and (b) verify whether the text data from a data provider has been used to train an LLM (data provenance). In this paper, we show that both problems can be solved by watermarking, i.e., by enabling an LLM to generate synthetic texts with embedded watermarks that contain information about their source(s). We identify the key properties of such watermarking frameworks (e.g., source attribution accuracy, robustness against adversaries), and propose a WAtermarking for Source Attribution (WASA) framework that satisfies these key properties due to our algorithmic designs. Our WASA framework enables an LLM to learn an accurate mapping from the texts of different data providers to their corresponding unique watermarks, which sets the foundation for effective source attribution (and hence data provenance). Extensive empirical evaluations show that our WASA framework achieves effective source attribution and data provenance.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 7512