19 upvotes
Meta Expert: Cross-Layer Expert Sharing for Storage- and Retrieval-Efficient MoE
What is Meta Expert: Cross-Layer Expert Sharing for Storage- and Retrieval-Efficient MoE?
As model parameters scale up, sparsely activated Mixture-of-Experts (MoE) approaches have become essential for model scaling. However, mainstream models currently adopt hierarchical architectures, where experts in different layers, although topologically independent, exhibit substantial similarity a
The numbers
| Project | Meta Expert: Cross-Layer Expert Sharing for Storage- and Retrieval-Efficient MoE |
|---|---|
| Type | Paper |
| Upvotes | 19 |
| First spotted by Early Signal | 23 Oct 2026 04:58 UTC |
| Seen on | Papers |
Open Meta Expert: Cross-Layer Expert Sharing for Storage- and Retrieval-Efficient MoE →