Publicly Detectable Watermarking for Language Models

ePrint Report: Publicly Detectable Watermarking for Language Models

Jaiden Fairoze, Sanjam Garg, Somesh Jha, Saeed Mahloujifar, Mohammad Mahmoody, Mingyuan Wang

We construct the first provable watermarking scheme for language models with public detectability or verifiability: we use a private key for watermarking and a public key for watermark detection. Our protocol is the first watermarking scheme that does not embed a statistical signal in generated text. Rather, we directly embed a publicly-verifiable cryptographic signature using a form of rejection sampling. …

detection eprint report key language language models private private key protocol public public key report wang watermarking