CryptoDB
Somesh Jha
Publications
Year
Venue
Title
2025
CIC
Publicly-Detectable Watermarking for Language Models
Abstract
<p> 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. </p>
Coauthors
- Jaiden Fairoze (1)
- Sanjam Garg (1)
- Somesh Jha (1)
- Saeed Mahloujifar (1)
- Mohammad Mahmoody (1)
- Mingyuan Wang (1)