search

richardbaihe / paperreading

NLP papers
MIT License
2 stars 0 forks source link

Arixv 2020|Efficient Content-Based Sparse Attention with Routing Transformers #29

Closed richardbaihe closed 4 years ago

richardbaihe commented 4 years ago

https://arxiv.org/abs/2003.05997

Cluster-based self-attention with the K-means algorithm.