Transformers
Fastformer
Introduced by Wu et al. in Fastformer: Additive Attention Can Be All You NeedFastformer is an type of Transformer which uses additive attention as a building block. Instead of modeling the pair-wise interactions between tokens, additive attention is used to model global contexts, and then each token representation is further transformed based on its interaction with global context representations.
Source: Fastformer: Additive Attention Can Be All You Need
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Mamba | 1 | 25.00% |
| Self-Supervised Learning | 1 | 25.00% |
| Text Classification | 1 | 25.00% |
| Text Summarization | 1 | 25.00% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
