PLAID：一个高效的后期交互检索引擎
PLAID: An Efficient Engine for Late Interaction Retrieval
摘要：
预训练的语言模型是跨多种信息检索（IR）范例的越来越重要的组件。后期交互是由ColBERT模型引入的，最近在ColBERTv2中得到了改进，是一种流行的范式，在许多基准测试中都具有最先进的地位。为了显著加快后期交互的搜索延迟，我们引入了性能优化的后期交互驱动程序（PLAID）。在不影响质量的情况下，PLAID使用一种新颖的质心交互机制快速消除低得分通道，该机制将每个通道视为一个轻量级的质心袋。PLAID在高度优化的引擎中使用质心交互和质心修剪（一种用于稀疏化质心包的机制），以将后期交互搜索延迟在GPU上减少多达7个$\times$，在CPU上减少多达45个$\times$，同时继续提供最先进的检索质量。这使得带有ColBERTv2的PLAID引擎能够在GPU上实现数十毫秒的延迟，在CPU上实现数十或数百毫秒的大规模延迟，即使在我们评估的最大规模（140 M通道）下也是如此。
英文摘要：
Pre-trained language models are increasingly important components across multiple information retrieval (IR) paradigms. Late interaction, introduced with the ColBERT model and recently refined in ColBERTv2, is a popular paradigm that holds state-of-the-art status across many benchmarks. To dramatically speed up the search latency of late interaction, we introduce the Performance-optimized Late Interaction Driver (PLAID). Without impacting quality, PLAID swiftly eliminates low-scoring passages using a novel centroid interaction mechanism that treats every passage as a lightweight bag of centroids. PLAID uses centroid interaction as well as centroid pruning, a mechanism for sparsifying the bag of centroids, within a highly-optimized engine to reduce late interaction search latency by up to 7$\times$ on a GPU and 45$\times$ on a CPU against vanilla ColBERTv2, while continuing to deliver state-of-the-art retrieval quality. This allows the PLAID engine with ColBERTv2 to achieve latency of tens of milliseconds on a GPU and tens or just few hundreds of milliseconds on a CPU at large scale, even at the largest scales we evaluate with 140M passages.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
