Skip to content

its-emile/watermarks

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 

Repository files navigation

Watermarks for LLMs

Python experiments to add, and detect watermarks in natural language text. I originally outlined this idea in a suggestion to OpenAI in September 2022.

With my initial algorithm, some of the original inspiration came from markov chain based text steganography by H Moraldo, 2014 using n-grams (before LLMs offered reliable probabilities for next token)

Relevant (but not used)

  • Kirchenbauer et al, 2023
  • A more robust implementation (including Reed Solomon error correction) is exciting, available here

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published