LLMs can now be used for efficient cryptography
LLMs can now be used for efficient cryptography
arxiv.org
LLMs can hide text in other text of the same length
A meaningful text can be hidden inside another, completely different yet still coherent and plausible, text of the same length. For example, a tweet containing a harsh political critique could be embe...
This paper describes a method by which an LLM can be used to hide a text, list, or programming code using a prompt with the right key, by masking it as an output text of the same token size. The same method can also be used for jailbreaks, i.e. circumventing censorship of the answers given by the model