# MixedNUTS: Training-Free Accuracy-Robustness Balance via Nonlinearly Mixed Classifiers
This is the official model repository of the preprint paper \
*[MixedNUTS: Training-Free Accuracy-Robustness Balance via Nonlinearly Mixed Classifiers](https://arxiv.org/abs/2402.02263)* \
by [Yatong Bai](https://bai-yt.github.io), [Mo Zhou](https://cdluminate.github.io), [Vishal M. Patel](https://engineering.jhu.edu/faculty/vishal-patel), and [Somayeh Sojoudi](https://www2.eecs.berkeley.edu/Faculty/Homepages/sojoudi.html).
**TL;DR:** MixedNUTS balances clean data classification accuracy and adversarial robustness without additional training
via a mixed classifier with nonlinear base model logit transformations.
Here, we provide the download links to the standard base classifiers used in the main results.
| Dataset | Link |
|-----------|-------|
| CIFAR-10 | [Download](http://172.233.227.28/base_models/cifar10/cifar10_std_rn152.pt) |
| CIFAR-100 | [Download](http://172.233.227.28/base_m…
