Grouter: Decoupling Routing from Representation for Accelerated MoE Training
Paper • 2603.06626 • Published • 1
Yuqi Xu
GreenAwoe
AI & ML interests
I'm interesting in all novel method about ml/dl.
Recent Activity
authored a paper 6 days ago
Grouter: Decoupling Routing from Representation for Accelerated MoE Training upvoted a paper 6 days ago
Grouter: Decoupling Routing from Representation for Accelerated MoE Training upvoted a paper about 1 month ago
Near-Future Policy OptimizationOrganizations
None yet