MICO: Selective Search with Mutual Information Co-training

TL;DR

MICO introduces a mutual information co-training framework that enhances selective search by clustering documents and routing queries efficiently, significantly improving retrieval performance with minimal supervision.

Contribution

The paper presents MICO, a novel co-training framework that improves selective search in large-scale systems using minimal supervision from search logs.

Findings

MICO significantly improves selective search performance.

MICO outperforms existing baselines in empirical tests.

MICO effectively routes unseen queries to relevant clusters.

Abstract

In contrast to traditional exhaustive search, selective search first clusters documents into several groups before all the documents are searched exhaustively by a query, to limit the search executed within one group or only a few groups. Selective search is designed to reduce the latency and computation in modern large-scale search systems. In this study, we propose MICO, a Mutual Information CO-training framework for selective search with minimal supervision using the search logs. After training, MICO does not only cluster the documents, but also routes unseen queries to the relevant clusters for efficient retrieval. In our empirical experiments, MICO significantly improves the performance on multiple metrics of selective search and outperforms a number of existing competitive baselines.