QCFuse: Query-Aware Cache Fusion via Compressed View for Efficient RAG Serving 事件

QCFuse: Query-Aware Cache Fusion via Compressed View for Efficient RAG Serving arXiv:2606.05875v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) improves large language model (LLM) answer quality by grounding generation in external evidence, but processing retrieved contexts makes the prefill stage a dominant serving cost. RAG cache fusion reduces this cost by reusing precomputed key-value (KV) caches for retrieved chunks and selectively recomputing tokens under the current p