Abstract
This paper introduces PAG-a novel optimization and decoding approach that guides autoregressive generation of document identifiers in generative retrieval models through simultaneous decoding. To this aim, PAG constructs a set-based and sequential identifier for each document. Motivated by the bag-of-words assumption in information retrieval, the set-based identifier is built on lexical tokens. The sequential identifier, on the other hand, is obtained via quantizing relevance-based representations of documents. Extensive experiments on MSMARCO and TREC Deep Learning Track data reveal that PAG outperforms the state-of-the-art generative retrieval model by a large margin (e.g., 15.6% MRR improvements on MS MARCO), while achieving 22x speed up in terms of query latency.
Abstract (translated)
本文介绍了一种名为PAG的新优化和解码方法,该方法通过同时解码在生成检索模型的自回归过程中指导文档标识符的生成。为实现这一目标,PAG构建了一个基于集合的文档标识符。受到信息检索中语料库假设的启发,基于词汇的标识符是通过量化相关性表示的文档。另一方面,顺序标识符则是通过量化文档的相关性表示获得的。在MSMARCO和TREC Deep Learning Track数据集的广泛实验中,PAG在生成检索模型的最先进实现方面显著超过了现有水平(例如,在MS MARCO数据集上提高了15.6%的MRR),同时将查询延迟速度提高了22倍。
URL
https://arxiv.org/abs/2404.14600