Verified80% confidenceFactTime unknown
Google systematized the Mixture of Experts architecture design in its 2017 Sparsely-Gated MoE paper.
6
Sources
80%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
小米MiMo V2.5 Pro深度实测：代码、3D、SVG生成能力全面评测
bilibiliovo_ojo
Related Entities
Related Claims
Unverified谷歌于2017年通过Shazeer等人的Sparsely-Gated MoE论文将MoE规模化至语言模型78% similarVerifiedThe Mixture of Experts (MoE) architecture was scaled up within the Transformer framework by Google Brain in 2017; its core idea is to group model parameters into multiple expert sub-networks, with only a small subset activated during each inference.75% similarUnverifiedGoogle的论文和NVIDIA NeMo框架都提供了工业级的上下文偏置实现74% similarUnverifiedMoE架构由 Google Brain 于 2017 年提出，GPT-4、Mixtral 8x7B 等知名模型均采用该架构73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/20622API
curl https://kongchang.com/api/v1/knowledge/claims/20622MCP
get_claim(id=20622)