Multi-Bit Distortion-Free Watermarking for Large Language Models

arXiv:2402.16578v1 Announce Type: new
Abstract: 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 …

abstract adversarial ai-generated text arxiv cs.cl cs.lg detection distribution free generated human language language models large language large language models quality text type watermark watermarking