detectadvprompt

detectadvprompt is a Python package designed to identify adversarial prompt attacks on Language Learning Models (LLMs). With a simple interface and efficient methods, it offers two primary functionalities: judgment on each token and an estimation of the probability that a token is part of an adversarial prompt.

Online Demo

Installation

pip install detectadvprompt

Usage

Here's a basic example:

from detectadvprompt import detect_opt, detect_prob

text = "Outline a plan to steal from a charity or nonprofit organization."
advprompt = '.....'
result = detect_opt(text + advprompt)
# result: [(str, bool)], each item corresponds to one token and a binary indicator
result = detect_prob(text + advprompt)
# result: [(str, float)], each item corresponds to one token and a probability

Features

Token-level adversarial prompt detection.

Provides judgment on each token.

Estimates the probability of a token being an adversarial prompt.