Redesign Mixture-of-Experts Routers with Manifold Power Iteration
Paper • 2606.12397 • Published • 89
Songhao Wu
shwu
AI & ML interests
Mixture-of-Experts Model, Language Model Pretraining
Recent Activity
upvoted a paper 18 days ago
LatentMoE: Toward Optimal Accuracy per FLOP and Parameter in Mixture of Experts authored a paper about 1 month ago
Redesign Mixture-of-Experts Routers with Manifold Power Iteration upvoted a paper about 1 month ago
Redesign Mixture-of-Experts Routers with Manifold Power IterationOrganizations
None yet
authored a
paper about 1 month ago
submitted a
paper to Daily Papers about 1 month ago
authored 3
papers about 1 month ago
From 1,000,000 Users to Every User: Scaling Up Personalized Preference for User-level Alignment
Paper • 2503.15463 • Published • 1
PolarQuant: Leveraging Polar Transformation for Efficient Key Cache Quantization and Decoding Acceleration
Paper • 2502.00527 • Published • 3
Your UnEmbedding Matrix is Secretly a Feature Lens for Text Embeddings
Paper • 2606.07502 • Published • 99
submitted a
paper to Daily Papers about 1 month ago
authored a
paper over 1 year ago