mirbostani/RobustQA
RobustQA: A Framework for Adversarial Text Generation Analysis on Question Answering Systems
This framework helps developers and researchers understand how robust their Question Answering (QA) systems are against intentionally misleading text. It takes an existing QA model and a dataset, then applies various attack strategies to subtly alter the input text. The output shows how easily the QA system can be fooled and provides metrics on performance degradation, helping improve the model's reliability. This is for AI/ML engineers building or evaluating natural language processing systems.
No commits in the last 6 months.
Use this if you are developing or deploying a question answering system and need to assess its vulnerability to adversarial text inputs.
Not ideal if you are looking for a general-purpose tool for text classification robustness or a non-technical end-user application.
Stars
8
Forks
2
Language
Python
License
MIT
Category
Last pushed
Dec 10, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/mirbostani/RobustQA"
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...