Unverified50% confidenceFactExact time
KV-Surgeon is an offline KV-cache compression method that uses calibration data to account for interactions among cache entries, deriving a quadratic surrogate model to make joint greedy deletion decisions and performing compensation among the remaining entries.
1
Sources
50%
Confidence
Long-term
Relevance
10/20/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedKV-Surgeon's compensation term can be computed directly via the closed-form solution of the surrogate model without additional forward passes, so its computational cost is far lower than exhaustive or iterative optimization.76% similarVerifiedThe KV cache is a core data structure in Transformer autoregressive inference, and its size is approximately 2 × number of layers × number of attention heads × head dimension × sequence length × precision bytes.71% similarVerifiedKV Cache是Transformer架构的核心推理优化机制，其显存占用量会随序列长度和并发批次线性增长68% similarUnverifiedKV Cache 的核心思想是缓存前面已计算的中间结果，避免每步从头计算68% similarUnverifiedOperational tuning experience in areas such as KV Cache management, batch scheduling, and VRAM allocation is almost entirely decoupled from specific model versions.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1073858API
curl https://kongchang.com/api/v1/knowledge/claims/1073858MCP
get_claim(id=1073858)