Unverified50% confidenceFactExact time
Sparse Transformer was proposed by OpenAI in 2019, replacing full attention with fixed sparse patterns.
1
Sources
50%
Confidence
Long-term
Relevance
8/31/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOpenAI于2024年推出Prompt Caching功能，底层依托Transformer架构中的KV Cache实现66% similarUnverifiedThe Universal Transformer was proposed in 2018 and repeatedly applies the same Transformer block to the input sequence.63% similarUnverifiedDomain randomization was first systematically applied by OpenAI in their 2017 robotic hand research63% similarUnverifiedLlama系列由Meta AI于2023年以开源形式发布，采用Transformer Decoder架构，通过RoPE和GQA等优化降低推理显存占用62% similarUnverifiedSwitch Transformer, proposed in 2021, simplified the number of experts routed to per token to 1.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/828091API
curl https://kongchang.com/api/v1/knowledge/claims/828091MCP
get_claim(id=828091)