gallilmaimon/LUNATC
This is the official implementation of "A Universal Adversarial Policy for Text Classifiers", Neural Networks (2022), https://doi.org/10.1016/j.neunet.2022.06.018
This project helps machine learning researchers evaluate the robustness of text classifiers against adversarial attacks. It takes an existing text classification model and a dataset as input. It then generates slightly modified text inputs that aim to fool the classifier while remaining semantically similar to the original, producing metrics on how susceptible the classifier is to these attacks. Researchers in natural language processing or machine learning who are focused on model security and resilience would find this useful.
No commits in the last 6 months.
Use this if you are a machine learning researcher or engineer developing text classifiers and need to thoroughly test their vulnerability to sophisticated adversarial text attacks.
Not ideal if you are an end-user looking for a simple tool to cleanse or preprocess text for classification, or if you are not familiar with machine learning model evaluation.
Stars
9
Forks
—
Language
Python
License
MIT
Category
Last pushed
Aug 23, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/gallilmaimon/LUNATC"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
thunlp/OpenAttack
An Open-Source Package for Textual Adversarial Attack.
thunlp/TAADpapers
Must-read Papers on Textual Adversarial Attack and Defense
jind11/TextFooler
A Model for Natural Language Attack on Text Classification and Inference
thunlp/OpenBackdoor
An open-source toolkit for textual backdoor attack and defense (NeurIPS 2022 D&B, Spotlight)
thunlp/HiddenKiller
Code and data of the ACL-IJCNLP 2021 paper "Hidden Killer: Invisible Textual Backdoor Attacks...