Unverified50% confidenceFactExact time
Switch Transformer was proposed by William Fedus et al. in 2021, and was the first to scale MoE to the trillion-parameter range
1
Sources
50%
Confidence
Long-term
Relevance
8/15/2026
First Seen
Sources
Gemini Flash文档处理能力被低估？基准测试验证用户真实体验
Related Claims
VerifiedGoogle's Switch Transformer, released in 2021, applied MoE at scale to Transformer language models, expanding model capacity to the trillion-parameter level.73% similarUnverifiedSwitch Transformer, proposed in 2021, simplified the number of experts routed to per token to 1.71% similarUnverifiedSwitch Transformer将专家数量扩展至2048个，2022年谷歌GLaM模型进一步验证了MoE在多任务泛化上的优势71% similarUnverifiedTransformer于2017年提出后被快速引入时序预测领域64% similarUnverifiedThe Universal Transformer was proposed in 2018 and repeatedly applies the same Transformer block to the input sequence.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/749717API
curl https://kongchang.com/api/v1/knowledge/claims/749717MCP
get_claim(id=749717)