Unverified50% confidenceFactExact time
线性注意力思路由 Katharopoulos 等人在 2020 年论文《Transformers are RNNs》中系统化
1
Sources
50%
Confidence
Long-term
Relevance
7/22/2026
First Seen
Sources
Related Claims
UnverifiedThe field of mechanistic interpretability was largely founded by the Anthropic research team (especially Chris Olah) around 2021 through the "Transformer Circuits" series of papers.70% similarVerifiedTransformer最初由Vaswani等人在2017年的论文《Attention Is All You Need》中提出，用于机器翻译任务64% similarUnverified《Attention Is All You Need》论文在发表初期并未全面碾压所有基线，却奠定了Transformer时代的架构基础63% similarUnverifiedTransformer原始论文于2017年由Google Brain团队发表61% similarUnverified自我反思（self-reflection）由Shinn等人在2023年的Reflexion论文中系统化，该论文发表于NeurIPS 202360% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/592762API
curl https://kongchang.com/api/v1/knowledge/claims/592762MCP
get_claim(id=592762)