研究人员开发了 NeuRoute，一个新颖的基于神经网络的索引系统，专为高效的十亿级向量搜索而设计。该系统利用 Logit 引导神经网络路由，显著加快了索引构建速度，对于 BigANN-1B 和 Deep1B-1B 等数据集，构建过程可在不到一小时内完成。NeuRoute 在速度上超越了 OPQ+IVF-PQ 等现有方法，同时保持高召回率，实现了具有竞争力的准确率-吞吐量权衡。 AI
影响 这种新的索引方法可以加速依赖于高效相似性搜索的大规模 AI 应用的开发和部署。
排序理由 该集群包含一篇详细介绍向量搜索新方法的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2608.15438v1 Announce Type: cross Abstract: Building approximate nearest neighbor (ANN) indexes at billion scale is often dominated by expensive global clustering or graph construction, making time-to-index a first-order systems concern. We present NeuRoute, a learned hashi…
Building approximate nearest neighbor (ANN) indexes at billion scale is often dominated by expensive global clustering or graph construction, making time-to-index a first-order systems concern. We present NeuRoute, a learned hashing index that turns short binary codes into an eff…