Slim attention: cut your context memory in half without loss of accuracy -- K-cache is all you need for MHA
Paper • 2503.05840 • Published • 3
gräfics
graefics
·
AI & ML interests
None yet
Organizations
upvoted a paper 11 months ago
upvoted an article about 1 year ago
Article
Topic 33: Slim Attention, KArAt, XAttention and Multi-Token Attention Explained – What’s Really Changing in Transformers?
•
16
upvoted a paper about 1 year ago