The loss landscape of Large Language Models resemble basin!

LLM adversarial attacks

Official code implement of Robust Classification via a Single Diffusion Model

A package that achieves 95%+ transfer attack success rate against GPT-4

CGI Web Server implemented by Huanran

huanran detection attack research

Official code implement of "Your Diffusion Model is Secretly a Certifiably Robust Classifier"

Official pytorch implement of paper "Boosting Adversarial Attack with Similar Target"

a pytorch implement of Adversarially Adaptive Normalization for Single Domain Generalization

my implement of flow based generative models

My implementation of Rectified FLow

Official code implement of object detection part in "Rethinking Model Ensemble in Transfer-based Adversarial Attacks"

Expression Evaluation in c++

Some visualization function that I always use

Address the gradient vanishing in convnext

Implement the dataloader