Baran Hashemi

This is a WGAN-gp version of the LOGAN which I used a modified version of it for my work. It was really improving the result of my WGAN-gp, so I...

Pytorch_implementation

1

Is there any PyTorch implementation of YLG? tnx

Question about Over-Smoothing problem?

2

I wonder if you @lucidrains, have any suggestions for the over-smoothing problem with Transformer models (both decoder and encoder).

The dataset link for the zip file is not working. Could you replace it with a new one?

Update attend.py with Tropical Attention

4

Hi @lucidrains, I would like to draw your attention to the Tropical Attention mechanism [2505.17190](https://arxiv.org/abs/2505.17190). This approach maps input vectors into a tropical space, performs information routing using tropical idempotent...