Raibows/RMLM
RMLM: A Flexible Defense Framework for Proactively Mitigating Word-level Adversarial Attacks, ACL 2023.
This project helps machine learning engineers and NLP researchers defend their natural language processing models against "adversarial attacks." These attacks subtly change a few words in an input text to trick a model into making wrong predictions. This tool takes your existing text classification or sentiment analysis model and your dataset, then trains a defensive layer to make your model more robust and reliable against such manipulative inputs.
No commits in the last 6 months.
Use this if you are developing or deploying NLP models and are concerned about their vulnerability to word-level adversarial attacks.
Not ideal if you are looking for defenses against non-text-based adversarial attacks or concept drift, or if you need a plug-and-play solution without fine-tuning.
Stars
9
Forks
—
Language
Python
License
—
Category
Last pushed
Dec 03, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/Raibows/RMLM"
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...