DynaMoE: Dynamic Token-Level Expert Activation with Layer-Wise Adaptive Capacity for Mixture-of-Experts Neural Networks
Paper • 2603.01697 • Published • 2
Gökdeniz Gülmez PRO
Goekdeniz-Guelmez
AI & ML interests
Transformers / NLP / Multimodal / Realtime M2M / J.O.S.I.E.-Home
Recent Activity
upvoted an article 25 minutes ago
Continuous batching from first principles liked a dataset about 24 hours ago
Goekdeniz-Guelmez/MLX-Benchmark-V2 updated a dataset 1 day ago
Goekdeniz-Guelmez/MLX-Benchmark-V2Organizations
submitted a
paper to Daily Papers about 2 months ago
authored a
paper about 2 months ago