Unverified50% confidenceFactExact time
Product-key memory organizes memory as the Cartesian product of two sets of key vectors, with each memory slot indexed by the product of a pair of keys (k₁, k₂).
1
Sources
50%
Confidence
Long-term
Relevance
10/21/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe product-key structure expands memory capacity from O(N) to O(N²) while requiring only O(N) storage and computation.71% similarUnverifiedProduct-key memory requires computing the inner product of the query with all keys; when the number of keys is large, computational complexity rises significantly.69% similarUnverifiedThe formula for the number of KV cache bytes that must be moved per token is: kv_bytes_per_token = 2 * layers * kv_heads * head_dim * dtype_bytes, where the factor 2 corresponds to the two copies for Key and Value.67% similarUnverifiedKV Cache memory usage is proportional to the number of layers × number of attention heads × context length × precision; GQA reduces cache size by having multiple query heads share the same set of Key-Value heads, and this design has been adopted by the Llama 2/3 and Qwen series.67% similarUnverified差分启发式需要O(K×N)的存储空间（K个地标、N个节点），例如100万节点的图使用16个地标、每个距离值4字节存储，总内存约64MB65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1079931API
curl https://kongchang.com/api/v1/knowledge/claims/1079931MCP
get_claim(id=1079931)