What Drives Cross-lingual Ranking? Retrieval Approaches with Multilingual Language Models
It addresses performance challenges in cross-lingual search systems, particularly for under-resourced languages, by prioritizing semantic embeddings over translation-based methods, though it is incremental as it builds on existing retrieval techniques.
This work systematically evaluates cross-lingual information retrieval approaches, finding that dense retrieval models trained for CLIR outperform lexical methods and gain little from document translation, with improvements most pronounced for low-resource and cross-script language pairs.
Cross-lingual information retrieval (CLIR) enables access to multilingual knowledge but remains challenging due to disparities in resources, scripts, and weak cross-lingual semantic alignment in embedding models. Existing pipelines often rely on translation and monolingual retrieval heuristics, which add computational overhead and noise, degrading performance. This work systematically evaluates four intervention types, namely document translation, multilingual dense retrieval with pretrained encoders, contrastive learning at word, phrase, and query-document levels, and cross-encoder re-ranking, across three benchmark datasets. We find that dense retrieval models trained specifically for CLIR consistently outperform lexical matching methods and derive little benefit from document translation. Contrastive learning mitigates language biases and yields substantial improvements for encoders with weak initial alignment, and re-ranking can be effective, but depends on the quality of the cross-encoder training data. Although high-resource languages still dominate overall performance, gains over lexical and document-translated baselines are most pronounced for low-resource and cross-script pairs. These findings indicate that cross-lingual search systems should prioritise semantic multilingual embeddings and targeted learning-based alignment over translation-based pipelines, particularly for cross-script and under-resourced languages.