Popular GitHub Repositories for Boundary Attack

Adversarial Robustness Toolbox (ART) - Python Library for Machine Learning Security - Evasion, Poisoning, Extraction, Inference - Red and Blue Teams

Master Federated Learning in 2 Hours—Run It on Your PC!

FedML - The Research and Production Integrated Federated Learning Library: https://fedml.ai

Must-read Papers on Textual Adversarial Attack and Defense

Developer friendly Natural Language Processing ?

Keras implementation of "One pixel attack for fooling deep neural networks" using differential evolution on Cifar10 and ImageNet

It's not AI that takes away your job, but the people who master the use of AI tools. The most deadly attack is a dimension-reducing strike: destroying you has nothing to do with you - from "The Three-Body Problem". 中文说明： 抢走你工作的不是AI，而是掌握使用AI工具的人。 降维打击最为致命：毁灭你，与你何干《三体》

Chinese safety prompts for evaluating and improving the safety of LLMs. 中文安全prompts，用于评估和提升大模型的安全性。

A pytorch adversarial library for attack and defense methods on images and graphs

Tensorflow implementation of "BEGAN: Boundary Equilibrium Generative Adversarial Networks"

A curated list of adversarial attacks and defenses papers on graph-structured data.

A library for solving differential equations using neural networks based on PyTorch, used by multiple research groups around the world, including at Harvard IACS.

An Open-Source Package for Textual Adversarial Attack.

Privacy Meter: An open-source library to audit data privacy in statistical and machine learning algorithms.

Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities. arXiv:2408.07666.

Deep Occlusion-Aware Instance Segmentation with Overlapping BiLayers [CVPR 2021]

A Model for Natural Language Attack on Text Classification and Inference

A Python library for adversarial machine learning focusing on benchmarking adversarial robustness.

A script for credentials-based attack surface enumeration and general reconnaissance of massive networks

A unified, easily extensible repository for LLM unlearning benchmarks (TOFU, MUSE) - enabling new datasets, evaluations, methods, and tasks.