Cross-Lingual Relevance Transfer for Document Retrieval

TL;DR

This paper demonstrates that multilingual BERT can effectively transfer relevance models across languages, improving document retrieval performance in multiple languages without additional processing.

Contribution

It shows that multilingual BERT trained on English data enhances both monolingual and cross-lingual retrieval in diverse languages, leveraging zero-shot transfer capabilities.

Findings

Improved ranking quality in five diverse languages.

Effective zero-shot cross-lingual relevance transfer.

No special processing needed for multilingual retrieval.

Abstract

Recent work has shown the surprising ability of multi-lingual BERT to serve as a zero-shot cross-lingual transfer model for a number of language processing tasks. We combine this finding with a similarly-recently proposal on sentence-level relevance modeling for document retrieval to demonstrate the ability of multi-lingual BERT to transfer models of relevance across languages. Experiments on test collections in five different languages from diverse language families (Chinese, Arabic, French, Hindi, and Bengali) show that models trained with English data improve ranking quality, without any special processing, both for (non-English) mono-lingual retrieval as well as cross-lingual retrieval.