AI RESEARCH
ButterflyViT: 354$\times$ Expert Compression for Edge Vision Transformers
arXiv CS.CV
•
ArXi:2603.06746v1 Announce Type: new Deploying sparse Mixture of Experts(MoE) Vision Transformers remains a challenge due to linear expert memory scaling. Linear memory scaling s $N$ independent expert weight matrices requiring $\mathcal{O}(N_E \cdot d^2)$ memory, which exceeds edge devices memory budget. Current compression methods like quantization, pruning and low-rank factorization reduce constant factors but leave the scaling bottleneck unresolved. We