Nyströmformer replaces the self-attention in BERT-small and BERT-base using the proposed Nyström approximation. This reduces self-attention complexity to $O(n)$ and allows the Transformer to support longer sequences.
Source: Nyströmformer: A Nyström-Based Algorithm for Approximating Self-AttentionPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Classification | 1 | 16.67% |
Image Classification | 1 | 16.67% |
Natural Language Inference | 1 | 16.67% |
Question Answering | 1 | 16.67% |
Semantic Textual Similarity | 1 | 16.67% |
Sentiment Analysis | 1 | 16.67% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |