Text Retrieval with Multi-Stage Re-Ranking Models

TL;DR

This paper introduces a three-stage re-ranking approach combining model ensembles and larger language models to enhance text retrieval accuracy while maintaining efficient search speeds.

Contribution

It proposes a novel multi-stage re-ranking framework that balances retrieval accuracy and speed, utilizing ensembles and larger language models in a zero-shot setting.

Findings

Higher retrieval accuracy achieved

Reduced search speed decay

Effective in zero-shot evaluation

Abstract

The text retrieval is the task of retrieving similar documents to a search query, and it is important to improve retrieval accuracy while maintaining a certain level of retrieval speed. Existing studies have reported accuracy improvements using language models, but many of these do not take into account the reduction in search speed that comes with increased performance. In this study, we propose three-stage re-ranking model using model ensembles or larger language models to improve search accuracy while minimizing the search delay. We ranked the documents by BM25 and language models, and then re-ranks by a model ensemble or a larger language model for documents with high similarity to the query. In our experiments, we train the MiniLM language model on the MS-MARCO dataset and evaluate it in a zero-shot setting. Our proposed method achieves higher retrieval accuracy while reducing the retrieval speed decay.