{"type": "article", "title": "KVBoost: Chunk-Level Key-Value Cache Reuse with Deviation-Guided Recomputation for Efficient Large Language Model Inference", "publisher": "Web Pulse", "url": "https://wpnews.pro/news/kvboost-chunk-level-key-value-cache-reuse-with-deviation-guided-recomputation", "original_source": "https://arxiv.org/abs/2608.21362", "published": "2026-08-25T04:00:00+00:00", "accessed": "2026-08-25", "id": "kvboost-chunk-level-key-value-cache-reuse-with-deviation-guided-recomputation"}