MHLA: Restoring Expressivity of Linear Attention via Token-Level Multi-Head
Paper
•
2601.07832
•
Published
•
42
Abhishek Kumar
developerabhi01
AI & ML interests
Reinforcement Learning
Recent Activity
liked
a model
about 20 hours ago
cerebras/GLM-4.7-REAP-218B-A32B
liked
a model
about 22 hours ago
ekwek/Soprano-1.1-80M
upvoted
a
paper
1 day ago
MHLA: Restoring Expressivity of Linear Attention via Token-Level Multi-Head
Organizations
None yet
upvoted
a
paper
1 day ago