Unverified50% confidenceFactExact time
SPERA designs factorized temporal and spatial attention, with full attention modules periodically interleaved to reduce computational complexity.
1
Sources
50%
Confidence
Long-term
Relevance
10/9/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe 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.80% similarUnverifiedOptimization techniques such as Flash Attention and sparse attention can compress memory complexity to near-linear.77% similarUnverifiedSpeedup ratios for attention operators are often highly sensitive to specific shapes and scenarios, with the decode phase being memory-access-intensive and the prefill phase being compute-intensive.76% similarUnverifiedLinformer 将注意力矩阵低秩近似为 O(n) 复杂度，FlashAttention 通过重排计算顺序降低 GPU 内存带宽占用，Longformer 引入滑动窗口注意力74% similarUnverifiedSPERA introduces a spatial prior based on Legendre polynomials and integrates it into the attention mechanism to encode the geometric relationships among different scalp electrode arrangements.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/990523API
curl https://kongchang.com/api/v1/knowledge/claims/990523MCP
get_claim(id=990523)