1
GRKV: Global Regression for Training-Free KV Cache Compression in Long-Context LLMs
无需训练的全局回归方法，高效压缩长上下文LLM的KV缓存，兼顾速度与精度。
arXiv:2605.31105v1 Announce Type: new Abstract: Large language models (LLMs) with extended context lengths rely on the key-value (KV) cache to support…