Improving Semantic Proximity in Information Retrieval through Cross-Lingual Alignment

TL;DR

This paper identifies a problem in cross-lingual information retrieval where models prefer English documents over related non-English ones, and proposes a training strategy to improve cross-lingual alignment.

Contribution

The paper introduces a new training approach that enhances cross-lingual alignment in multilingual retrieval models using a small dataset.

Findings

Significant improvement in cross-lingual retrieval performance

Reduction of English document bias in retrieval results

Enhanced alignment capabilities across multiple multilingual models

Abstract

With the increasing accessibility and utilization of multilingual documents, Cross-Lingual Information Retrieval (CLIR) has emerged as an important research area. Conventionally, CLIR tasks have been conducted under settings where the language of documents differs from that of queries, and typically, the documents are composed in a single coherent language. In this paper, we highlight that in such a setting, the cross-lingual alignment capability may not be evaluated adequately. Specifically, we observe that, in a document pool where English documents coexist with another language, most multilingual retrievers tend to prioritize unrelated English documents over the related document written in the same language as the query. To rigorously analyze and quantify this phenomenon, we introduce various scenarios and metrics designed to evaluate the cross-lingual alignment performance of multilingual retrieval models. Furthermore, to improve cross-lingual performance under these challenging conditions, we propose a novel training strategy aimed at enhancing cross-lingual alignment. Using only a small dataset consisting of 2.8k samples, our method significantly improves the cross-lingual retrieval performance while simultaneously mitigating the English inclination problem. Extensive analyses demonstrate that the proposed method substantially enhances the cross-lingual alignment capabilities of most multilingual embedding models.