 |
Softmax vs. Linear Attention in the Large-Prompt Regime
1 : Laboratoire de Mathématiques d'Orsay
Université Paris-Saclay, Centre National de la Recherche Scientifique, Centre National de la Recherche Scientifique : UMR8628
| Subject : | : | Minisymposium talk (upon invitation only) |
| Topics | : | MS: Mathematical aspects of deep learning (upon invitation only) |
| Keywords | : | Attention layers ; transformers ; Training dynamics ; In ; context learning |
| PDF version | : |  PDF version |
