让检索模型可变大小，兼顾效率与效果
The Matryoshka Hypencoder
- 用多层嵌套结构让查询网络支持不同规模
- 域内性能相当，参数少7倍，域外参数减半
- 适合需要高效部署的检索系统
Hypencoder 是一种新型检索方法，将查询编码为浅层神经网络（Q-Nets），用于估计预计算文档嵌入的相关性。受俄罗斯套娃表征学习启发，本文提出可支持多种尺寸 Q-Nets 的「马特里什卡 Hypencoder」，实现效果与效率的权衡。实验表明，该方法在域内达到相近效果，但主动参数减少约7倍；域外则主动参数减半，评分吞吐量提升1.6至3.4倍。这项工作为 Hypencoder 的实际部署铺平了道路。
原文摘要 · Abstract (English)
The Hypencoder is a recently-proposed retrieval approach that encodes queries as shallow neural networks ("Q-Nets") that estimate relevance over pre-computed document embeddings. Inspired by Matryoshka Representation Learning, we show that the Hypencoder can be extended to support multiple sizes of Q-Nets, allowing trade-offs between effectiveness and efficiency when deployed. We find that this "Matryoshka Hypencoder" achieves comparable in-domain effectiveness with approximately 7x fewer active parameters in-domain and half as many active parameters out-of-domain, which corresponds to a 1.6-3.4x increase in scoring throughput. This work paves the way for practical deployment of Hypencoders.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。