Methods for watermarking large language models have been proposed that distinguish AI-generated text from human-generated text by slightly altering the model output distribution, but they also distort the quality of the text, exposing the watermark to adversarial detection. More recently, distortion-free watermarking methods were proposed that require a secret key to detect the watermark. The prior methods generally embed zero-bit watermarks that do not provide additional information beyond tagging a text as being AI-generated. We extend an existing zero-bit distortion-free watermarking method by embedding multiple bits of meta-information as part of the watermark. We also develop a computationally efficient decoder that extracts the embedded information from the watermark with low bit error rate.
@article{arxiv.2402.16578,
title = {Multi-Bit Distortion-Free Watermarking for Large Language Models},
author = {Massieh Kordi Boroujeny and Ya Jiang and Kai Zeng and Brian Mark},
journal= {arXiv preprint arXiv:2402.16578},
year = {2024}
}