thunlp/Advbench
Code and data of the EMNLP 2022 paper "Why Should Adversarial Perturbations be Imperceptible? Rethink the Research Paradigm in Adversarial NLP".
This project helps security researchers and natural language processing (NLP) specialists evaluate the robustness of NLP systems against malicious text attacks. It provides a collection of security-oriented datasets and methods to simulate realistic adversarial attacks. Researchers can input these datasets and attack methods to test NLP models, receiving insights into their vulnerabilities in security scenarios.
No commits in the last 6 months.
Use this if you are an NLP security researcher aiming to develop or evaluate NLP models that are resilient to real-world textual adversarial attacks.
Not ideal if you are looking for a general-purpose NLP library for tasks like sentiment analysis or machine translation, as its focus is specifically on adversarial robustness in security.
Stars
72
Forks
9
Language
Python
License
MIT
Category
Last pushed
Feb 19, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/thunlp/Advbench"
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...