You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardexpand all lines: README.md
+2
Original file line number
Diff line number
Diff line change
@@ -1,5 +1,7 @@
1
1
# Spurious Features Everywhere - Large-Scale Detection of Harmful Spurious Features in ImageNet
2
2
3
+
***Coming soon: 60 additional spurious features for 60 new classes for the Spurious Imagenet dataset***
4
+
3
5
This repository will contain the code for our paper [Spurious Features Everywhere - Large-Scale Detection of Harmful Spurious Features in ImageNet](https://arxiv.org/abs/2212.04871) including the *Spurious ImageNet* dataset.
4
6
5
7
In this paper, we develop a framework that allows us to systematically identify spurious features in large datasets like ImageNet. It is based on our neural PCA components and their visualization.
0 commit comments