Unverified50% confidenceFactExact time
原始 Self-Attention 的计算量随序列长度呈平方级增长，Anthropic 通过优化位置编码（如 ALiBi）等技术手段拓展了可用上下文
1
Sources
50%
Confidence
Long-term
Relevance
7/8/2026
First Seen
Sources
Related Claims
Unverified自注意力使任意两个位置之间的信息交互只需一步计算，但显存占用随序列长度平方增长81% similarVerifiedContext length and computational load grow in an approximately quadratic relationship, influenced by the attention mechanism.78% similarUnverifiedThe core operation of Self-Attention is to compute the dot-product similarity between query vectors and key vectors, then use this as weights to perform a weighted sum of the value vectors.76% similarUnverifiedAnthropic通过稀疏注意力和位置编码改进使得超长上下文在工程上变得可行74% similarUnverified自注意力机制的计算复杂度随序列长度呈近似二次方增长，序列长度为n时注意力矩阵计算量正比于n²74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/255057API
curl https://kongchang.com/api/v1/knowledge/claims/255057MCP
get_claim(id=255057)