CryptoDB
Publicly-Detectable Watermarking for Language Models
Authors: |
|
---|---|
Download: | |
Abstract: | We present a publicly-detectable watermarking scheme for LMs: the detection algorithm contains no secret information, and it is executable by anyone. We embed a publicly-verifiable cryptographic signature into LM output using rejection sampling and prove that this produces unforgeable and distortion-free (i.e., undetectable without access to the public key) text output. We make use of error-correction to overcome periods of low entropy, a barrier for all prior watermarking schemes. We implement our scheme and find that our formal claims are met in practice. |
BibTeX
@article{cic-2025-34924, title={Publicly-Detectable Watermarking for Language Models}, journal={cic}, publisher={International Association for Cryptologic Research}, volume={1, Issue 4}, url={https://cic.iacr.org/p/1/4/31}, doi={10.62056/ahmpdkp10}, author={Jaiden Fairoze and Sanjam Garg and Somesh Jha and Saeed Mahloujifar and Mohammad Mahmoody and Mingyuan Wang}, year=2025 }