Unverified50% confidenceFactExact time
MoE概念最早由Jacobs等人于1991年提出，谷歌2017年将其规模化至语言模型，Mistral的Mixtral系列将稀疏MoE带入开源社区
1
Sources
50%
Confidence
Long-term
Relevance
7/15/2026
First Seen
Sources
Related Claims
VerifiedThe core idea of MoE (Mixture of Experts) can be traced back to the early mixture of experts framework proposed by Jacobs et al. in 1991, and Google's Switch Transformer in 2021 brought it into the era of large-scale language models.76% similarUnverifiedMoE架构历史可追溯至1991年Jacobs等人的原始论文，Google于2017年将其应用于序列到序列模型并显著提升性能74% similarUnverifiedMoE架构最早由Google在2017年提出，后被Mistral的Mixtral系列和阿里的Qwen系列等广泛采用73% similarUnverified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72% similarUnverified混合专家架构(MoE)的思路源于1991年的学术研究，近年被Mistral、DeepSeek、Google(Gemini 1.5)等广泛采用71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/520787API
curl https://kongchang.com/api/v1/knowledge/claims/520787MCP
get_claim(id=520787)