Unverified50% confidenceFactExact time
标准的Scaled Dot-Product Attention随着序列长度N增加，每个位置能分配到的平均注意力权重以1/N的速率稀释
1
Sources
50%
Confidence
Long-term
Relevance
7/19/2026
First Seen
Sources
Related Claims
Unverified标准 Scaled Dot-Product Attention 中 Q 与 K 做矩阵乘法得到 n×n 注意力分数矩阵，计算量为 O(n²·d)83% similarUnverified标准Attention的显存消耗随序列长度平方增长，Flash Attention将其压缩到线性增长78% 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n²75% similarUnverifiedThe computational cost of standard self-attention grows as O(n²) with sequence length, limiting the scale of context windows73% similarUnverifiedThe computational complexity of standard self-attention is quadratic with respect to sequence length, and achieving a context of one million tokens requires introducing techniques such as sparse attention, sliding window attention, and KV Cache compression.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/558190API
curl https://kongchang.com/api/v1/knowledge/claims/558190MCP
get_claim(id=558190)