ml-safety 开源项目推荐（26 个）

hendrycks /robustness

腐败与扰动的稳健性（ICLR 2019）

hendrycks /natural-adv-examples

更难的ImageNet测试集（CVPR 2021）

hendrycks /outlier-exposure

Deep Anomaly Detection with Outlier Exposure (ICLR 2019)

PacificAI /langtest

Deliver safe & effective language models

agencyenterprise /PromptInject

PromptInject 是一个模块化组装提示的框架，旨在定量分析 LLMs 对对抗性提示攻击的鲁棒性。🏆 最佳论文奖 @ NeurIPS 机器学习安全研讨会 2022

hendrycks /ethics

Aligning AI With Shared Human Values (ICLR 2021)

hendrycks /imagenet-r

ImageNet-R(endition) and DeepAugment (ICCV 2021)

hendrycks /ss-ood

Self-Supervised Learning for OOD Detection (NeurIPS 2019)

Giskard-AI /awesome-ai-safety

📚 A curated list of papers & technical articles on AI Quality & Safety

jiachens /ModelNet40-C

Repo for "Benchmarking Robustness of 3D Point Cloud Recognition against Common Corruptions" https://arxiv.org/abs/2201.12296

hendrycks /anomaly-seg

The Combined Anomalous Object Segmentation (CAOS) Benchmark

hendrycks /pre-training

Pre-Training Buys Better Robustness and Uncertainty Estimates (ICML 2019)

YyzHarry /ME-Net

[ICML 2019] ME-Net: Towards Effective Adversarial Robustness with Matrix Estimation

hendrycks /jiminy-cricket

Jiminy Cricket Environment (NeurIPS 2021)

yaodongyu /ProjNorm

Predicting Out-of-Distribution Error with the Projection Norm

moonwatcher-ai /moonwatcher

Evaluation & testing framework for computer vision models

Doleus /doleus

Build confidence in your AI with systematic slice-based testing

EzgiKorkmaz /AI-Safety

AAAI 2025 Tutorial on AI Safety

harsmac /MUFIACode

Code for the attack multiplicative filter attack MUFIA, from the paper "Frequency-based vulnerability analysis of deep learning models against image corruptions".

ArianeDlns /adv-AI-project

This repository contains the project for the Advanced AI course @CentraleSupélec