ReCalKV: Low-Rank KV Cache Compression via Head Reordering and Offline Calibration

Abstract: LLMs have achieved remarkable performance, yet their capability on long-context reasoning is often constrained by the excessive memory required to store the Key-Value (KV) cache. This makes KV cache compression an essential step toward enabling efficient long-context reasoning. Recent methods have explored reducing the hidden dimensions of the KV cache, but many introduce additional computation through projection layers or suffer from significant performance degradation under high compression ratios. To address these challenges, we propose ReCalKV, a post-training KV cache compression method that reduces the hidden dimensions of the KV cache. We develop distinct compression strategies for Keys and Values based on their different roles and varying importance in the attention mechanism. For Keys, we propose Head-wise Similarity-aware Reordering (HSR), which clusters similar heads and applies grouped SVD to the key projection matrix, reducing additional computation while preserving accuracy. For Values, we propose Offline Calibration and Matrix Fusion (OCMF) to preserve accuracy without extra computational overhead. Experiments show that ReCalKV outperforms existing low-rank compression methods, achieving high compression ratios with minimal performance loss. The code and models will be available at: https://github.com/XIANGLONGYAN/ReCalKV.