1
AnchorKV: Anchor-Residual KV Cache Compression
锚点加残差，KV缓存压缩新思路，兼顾低内存与高保真，值得一看。
arXiv:2608.02901v1 Announce Type: new Abstract: The key-value (KV) cache is the primary memory bottleneck in long-context LLM inference. Existing appr…
锚点加残差，KV缓存压缩新思路，兼顾低内存与高保真，值得一看。
arXiv:2608.02901v1 Announce Type: new Abstract: The key-value (KV) cache is the primary memory bottleneck in long-context LLM inference. Existing appr…