Link, Synthesize, Retrieve: Universal Document Linking for Zero-Shot Information Retrieval

TL;DR

This paper introduces a Universal Document Linking algorithm that improves zero-shot information retrieval by linking similar documents to generate better synthetic queries, demonstrating superior performance across diverse datasets.

Contribution

The paper presents a novel UDL algorithm that enhances zero-shot IR by linking documents using entropy and NER, outperforming existing methods.

Findings

UDL improves zero-shot IR performance across datasets.

UDL surpasses state-of-the-art methods in experiments.

Code for UDL is publicly available for reproducibility.

Abstract

Despite the recent advancements in information retrieval (IR), zero-shot IR remains a significant challenge, especially when dealing with new domains, languages, and newly-released use cases that lack historical query traffic from existing users. For such cases, it is common to use query augmentations followed by fine-tuning pre-trained models on the document data paired with synthetic queries. In this work, we propose a novel Universal Document Linking (UDL) algorithm, which links similar documents to enhance synthetic query generation across multiple datasets with different characteristics. UDL leverages entropy for the choice of similarity models and named entity recognition (NER) for the link decision of documents using similarity scores. Our empirical studies demonstrate the effectiveness and universality of the UDL across diverse datasets and IR models, surpassing state-of-the-art methods in zero-shot cases. The developed code for reproducibility is included in https://github.com/eoduself/UDL