1
Tying the Loop -- Tied Expert Layers in Mixture-of-Experts Language Models
MoE新范式：跨OLMoE、Qwen3等主流架构验证的Tied Expert Layers方法，预训练实验揭示连接专家层如何提升大模型效率。
arXiv:2606.16825v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) architectures efficiently scale Large Language Models (LLMs) by activating …