SOURCE-LINKED INTELLIGENCE

PuzzleKV: Page-Wise Low-Rank Decomposition for KV Cache Compression

arXiv · AI, language, vision and robotics · article · Aug 24, 2026 · UTC

Long-context inference in large language models (LLMs) is increasingly limited by the memory required for the key-value (KV) cache. KV cache compression addresses this problem by reducing the storage cost of previous tokens. Among existing approaches, low-rank compression is particularly attractive because it represents every token in reduced dimensions. Previous low-rank methods typically derive fixed projection spaces from model weights, construct fixed spaces from calibration activations, or construct a shared basis over a broad cache region. Such representations may not capture detailed bu

Read original source ↗ Open in workspace

recordType

paper

region

Global