A Watermark for Large Language Models. (arXiv:2301.10226v1 [cs.LG])

Web: http://arxiv.org/abs/2301.10226

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 …

language language models large