Update README.md

README.md

@@ -1,3 +1,87 @@
 ---
 license: apache-2.0
+task_categories:
+- image-classification
+size_categories:
+- n<1K
 ---
+## Neglected Free Lunch – Learning Image Classifiers Using Annotation Byproducts | [Paper](https://arxiv.org/abs/2303.17595)
+
+Dongyoon Han<sup>1*</sup>, Junsuk Choe<sup>2*</sup>, Seonghyeok Chun<sup>3</sup>, John Joon Young Chung<sup>4</sup>
+
+Minsuk Chang<sup>5</sup>, Sangdoo Yun<sup>1</sup>, Jean Y. Song<sup>6</sup>, Seong Joon Oh<sup>7&dagger;</sup>  
+
+<sub>\* Equal contribution</sub> <sub>&dagger;</sub> <sub> Corresponding author </sub>
+
+<sup>1</sup> <sub>NAVER AI LAB</sub> <sup>2</sup> <sub>Sogang University</sub> <sup>3</sup> <sub>Dante Company</sub> <sup>4</sup> <sub>University of Michigan</sub>  <sup>5</sup> <sub>NAVER AI LAB, currently at Google</sub>  <sup>6</sup> <sub>DGIST</sub>  <sup>7</sup> <sub>University of T&uuml;bingen</sub> 
+
+Supervised learning of image classifiers distills human knowledge into a parametric model *f* through pairs of images and corresponding labels (*X*,*Y*). We argue that this simple and widely used representation of human knowledge neglects rich auxiliary information from the annotation procedure, such as the time-series of mouse traces and clicks. 
+
+<p align=center>
+<img src="https://user-images.githubusercontent.com/7447092/203720567-dc6e1277-84d2-439c-a9f8-879e31c04e6f.png" alt="imagenet-byproduct-sample" width=500px />
+<p/>
+
+Our insight is that such **annotation byproducts** *Z* provide approximate human attention that weakly guides the model to focus on the foreground cues, reducing spurious correlations and discouraging shortcut learning. 
+
+We have created **ImageNet-AB** and **COCO-AB** to verify this:
+
+They are ImageNet and COCO training sets enriched with sample-wise annotation byproducts, collected by replicating the respective original annotation tasks. 
+
+We refer to the new paradigm of training models with annotation byproducts as **learning using annotation byproducts (LUAB)**. 
+
+<p align=center>
+<img src="https://user-images.githubusercontent.com/7447092/203721515-2aea133d-1a77-4463-8372-5f0e0dbe4d2d.png" alt="luab" width=500px />
+<p/>
+
+We show that a simple multitask loss for regressing *Z* together with *Y* already improves the generalisability and robustness of the learned models. Compared to the original supervised learning, LUAB does not require extra annotation costs.
+
+### Dataloader for ImageNet-AB and COCO-AB
+
+We provide example dataloaders for the annotation byproducts.
+
+* Dataloader for ImageNet-AB: [imagenet_dataloader.ipynb](imagenet_dataloader.ipynb)
+* Dataloader for COCO-AB: [coco_dataloader.ipynb](coco_dataloader.ipynb)
+
+
+### Annotation tools for ImageNet and COCO
+
+* Annotation tool for ImageNet: [github.com/naver-ai/imagenet-annotation-tool](https://github.com/naver-ai/imagenet-annotation-tool)
+* Annotation tool for COCO: [github.com/naver-ai/coco-annotation-tool](https://github.com/naver-ai/coco-annotation-tool)
+
+### License
+
+```
+MIT License
+
+Copyright (c) 2023-present NAVER Cloud Corp.
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+```
+
+### Citing our work
+
+```
+@article{han2023arxiv,
+  title = {Neglected Free Lunch – Learning Image Classifiers Using Annotation Byproducts},
+  author = {Han, Dongyoon and Choe, Junsuk and Chun, Seonghyeok and Chung, John Joon Young and Chang, Minsuk and Yun, Sangdoo and Song, Jean Y. and Oh, Seong Joon},
+  journal={arXiv preprint arXiv:2303.17595},
+  year = {2023}
+}
+```
+