1
ArborKV: Structure-Aware KV Cache Management for Scaling Tree-based LLM Reasoning
针对树状推理扩展难题，提出结构感知的KV缓存管理方法，显著提升LLM推理效率。
arXiv:2605.22106v1 Announce Type: new Abstract: Recent progress in LLM reasoning has increasingly shifted from single-pass generation to explicit sear…