Prompts have evil twins
2023-11-13Code Available1· sign in to hype
Rimon Melamed, Lucas H. McCabe, Tanay Wakhare, Yejin Kim, H. Howie Huang, Enric Boix-Adsera
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/rimon15/propaneOfficialIn paperpytorch★ 23
- github.com/rimon15/evil_twinsOfficialpytorch★ 23
Abstract
We discover that many natural-language prompts can be replaced by corresponding prompts that are unintelligible to humans but that provably elicit similar behavior in language models. We call these prompts "evil twins" because they are obfuscated and uninterpretable (evil), but at the same time mimic the functionality of the original natural-language prompts (twins). Remarkably, evil twins transfer between models. We find these prompts by solving a maximum-likelihood problem which has applications of independent interest.