Unverified50% confidenceFactExact time
Sliding-window attention limits the visible range of each position to a nearby fixed window, reducing computation but losing long-distance information.
1
Sources
50%
Confidence
Long-term
Relevance
11/2/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedLongformer采用滑动窗口注意力方案，允许每个token只关注局部窗口内的邻近token78% similarUnverifiedAs the number of tokens in the window grows, attention weights must be distributed among these tokens, and the average attention each individual token receives declines accordingly.77% similarVerifiedSliding Window Attention, adopted by models such as Mistral, limits each token's attention range to a fixed window.74% similarUnverified滑动窗口注意力（如Mistral采用的方案）通过限制每个token只关注固定窗口内的历史，将复杂度从O(n²)降至O(n·w)73% similarVerifiedFlash Attention significantly reduces GPU memory usage and raises the physically available window limit through IO-aware block computation.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1125743API
curl https://kongchang.com/api/v1/knowledge/claims/1125743MCP
get_claim(id=1125743)