Unverified50% confidenceFactExact time
The continuous form of a state space model is h'(t) = A·h(t) + B·x(t), with output y(t) = C·h(t) + D·x(t); after discretization it processes a sequence of length n with O(n) complexity.
1
Sources
50%
Confidence
Long-term
Relevance
10/17/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedState space models represented by Mamba compress historical information by maintaining a fixed-size hidden state, with both time and space complexity of O(1) during inference.69% similarUnverifiedDEER solves the computation of the entire sequence as a large system of equations in one shot through Newton-type fixed-point iteration, theoretically reducing complexity from O[T] to O[(log T)²].68% similarUnverified哈达玛矩阵满足正交性条件 H_n · H_n^T = n · I_n，使哈达玛变换可逆且不丢失信息66% similarUnverified典型的导纳模型可表达为 M·ẍ + B·ẋ + K·x = F_ext，其中M、B、K分别是期望的虚拟惯量、阻尼和刚度矩阵65% similarUnverifiedSSM 通过随时间演化的隐藏状态 h_t = A·h_{t-1} + B·x_t 压缩历史信息，输出由 h_t 线性映射得到，单步解码计算量与序列长度无关，显存占用恒定。65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1049991API
curl https://kongchang.com/api/v1/knowledge/claims/1049991MCP
get_claim(id=1049991)