Lean Attention: Hardware-Aware Scalable Attention Mechanism for the Decode-Phase of Transformers

Add code
May 17, 2024

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: