# ICLR 2021

- [ ] [HOW NEURAL NETWORKS EXTRAPOLATE: FROM FEEDFORWARD TO GRAPH NEURAL NETWORKS]
* [Paper](https://openreview.net/pdf?id=UH-cmocLJC)

- [ ] [RETHINKING ATTENTION WITH PERFORMERS]
* [Paper](https://openreview.net/pdf?id=Ua6zuk0WRH)
* [Code]
* [Pytorch](https://github.com/lucidrains/performer-pytorch)

- [ ] [Coupled Oscillatory Recurrent Neural Network (coRNN): An accurate and (gradient) stable architecture for learning long time dependencies]
* [Paper](https://openreview.net/pdf?id=F3s69XzWOia)
* [Code](https://github.com/tk-rusch/coRNN/tree/master/adding)