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.

本文介绍了 PAG，一种新的优化和解码方法，通过同时解码来指导生成式检索模型中文档标识符的自回归生成。PAG 为每个文档构建了一种基于集合和顺序的标识符，其中基于集合的标识符基于词汇标记建立，而顺序标识符则通过量化文档的基于相关性的表示获得。在 MSMARCO 和 TREC 深度学习赛道数据上进行的大量实验证明 PAG 在性能上超过了现有的生成式检索模型（例如，在 MS MARCO 上提高了 15.6% 的 MRR），同时查询延迟方面达到了 22 倍的加速。

生成式检索中的先行规划：通过同时解码引导自回归生成

Planning Ahead in Generative Retrieval: Guiding Autoregressive  Generation through Simultaneous Decoding

Generative retrieval (Wang et al., 2022; Tay et al., 2022) is a new approach
for end-to-end document retrieval that directly generates document identifiers
given an input query. Techniques for designing effective, high-quality document
IDs remain largely unexplored. We introduce ACID, in which each document's ID
is composed of abstractive keyphrases generated by a large language model,
rather than an integer ID sequence as done in past work. We compare our method
with the current state-of-the-art technique for ID generation, which produces
IDs through hierarchical clustering of document embeddings. We also examine
simpler methods to generate natural-language document IDs, including the naive
approach of using the first k words of each document as its ID or words with
high BM25 scores in that document. We show that using ACID improves top-10 and
top-20 accuracy by 15.6% and 14.4% (relative) respectively versus the
state-of-the-art baseline on the MSMARCO 100k retrieval task, and 4.4% and 4.0%
respectively on the Natural Questions 100k retrieval task. Our results
demonstrate the effectiveness of human-readable, natural-language IDs in
generative retrieval with LMs. The code for reproducing our results and the
keyword-augmented datasets will be released on formal publication.

通过生成式检索方法和使用大型语言模型生成抽象关键短语来构成文件的 ID，与现有的文档 ID 生成方法相比，我们的 ACID 方法在 MSMARCO 100k 检索任务和自然问题 100k 检索任务上分别使得前 10 和前 20 的准确率相对提高了 15.6％和 14.4％以及 4.4％和 4.0％。