Title | A Watermark for Large Language Models |
Publication Type | Journal Article |
Year of Publication | 2023 |
Authors | Kirchenbauer, J, Geiping, J, Wen, Y, Katz, J, Miers, I, Goldstein, T |
Date Published | 6/6/2023 |
Abstract | Potential harms of large language models can be mitigated by watermarking model output, i.e., embedding signals into generated text that are invisible to humans but algorithmically detectable from a short span of tokens. We propose a watermarking framework for proprietary language models. The watermark can be embedded with negligible impact on text quality, and can be detected using an efficient open-source algorithm without access to the language model API or parameters. The watermark works by selecting a randomized set of "green" tokens before a word is generated, and then softly promoting use of green tokens during sampling. We propose a statistical test for detecting the watermark with interpretable p-values, and derive an information-theoretic framework for analyzing the sensitivity of the watermark. We test the watermark using a multi-billion parameter model from the Open Pretrained Transformer (OPT) family, and discuss robustness and security. |
URL | https://arxiv.org/abs/2301.10226 |