Unverified50% confidenceFactExact time
DeltaNet belongs to linear attention/state space models, replacing the full-history KV storage in standard attention with an updatable recurrent state matrix, so computation and storage grow linearly rather than quadratically with sequence length.
1
Sources
50%
Confidence
Long-term
Relevance
10/20/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedDeltaNet is a variant of linear attention that replaces softmax attention with a delta update rule, reducing sequence processing complexity from O(n²) to O(n).78% similarUnverifiedGated DeltaNet makes memory usage during inference independent of context length by maintaining a fixed-size hidden state matrix.72% similarUnverifiedKDA (Kernelized Delta Attention) kernels the Delta rule, attempting to find a smoother interpolation between the recursive state and attention.67% similarUnverifiedMamba-2, Gated DeltaNet and KDA are three new state update mechanisms used to replace some attention layers, achieving efficient state memory through a recursive structure with linear complexity.66% similarUnverifiedKDA is a linear attention mechanism variant that uses a fixed-size recurrent state and does not retain KV cache.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1073328API
curl https://kongchang.com/api/v1/knowledge/claims/1073328MCP
get_claim(id=1073328)