Chawin Sitawarin's repositories
Adversarial-Examples-Reading-List
This is the reading list mainly on adversarial examples (attacks, defenses, etc.) I try to keep and update regularly.
adv-part-model
Code for a research paper "Part-Based Models Improve Adversarial Robustness" (ICLR 2023)
knn-defense
Adversarial Examples on KNN (and its neural network friends)
bagnet-adv
Exploring how BagNet can be used for interpretability and defending adversarial examples
ates-minimal
Improving Adversarial Robustness Through Progressive Hardening (AutoAttack test)
dknn_attack
Demonstrate attacks on kNN and Deep kNN
adversarial-nonparametrics
Adversarial Examples for Non-Parametric Methods: Attacks, Defenses and Large Sample Limits
noise-robust
Study effects of data augmentation with random noises to model boundary and robustness
chawins.github.io
A beautiful, simple, clean, and responsive Jekyll theme for academics
cs189_spring2023_notes
Mini-lecture notes on a variety of topics from UC Berkeley's Introduction to Machine Learning course (CS189/289A Spring 2023)
detect-pretrain-code
This repository provides an original implementation of Detecting Pretraining Data from Large Language Models by *Weijia Shi, *Anirudh Ajith, Mengzhou Xia, Yangsibo Huang, Daogao Liu , Terra Blevins , Danqi Chen , Luke Zettlemoyer.
ensemble_defense
Defending against adversarial examples with diverse ensemble
hinge_loss_nn
Experiments on neural networks with hinge loss
Mask2Former
Code release for "Masked-attention Mask Transformer for Universal Image Segmentation"
mimir
Python package for measuring memorization in LLMs.
panoptic_parts
This repository contains code and tools for reading, processing, evaluating on, and visualizing Panoptic Parts datasets. Moreover, it contains code for reproducing our CVPR 2021 paper results.
py-gutenberg
A Python library to access the Project Gutenberg API
pytorch-cifar
95.47% on CIFAR10 with PyTorch
savio-note
Some notes and instructions for using Savio server
TAP
TAP: An automated jailbreaking method for black-box LLMs
tensorflow2-cifar
95.76% on CIFAR-10 with TensorFlow2