Cross-lingual QA: A Key to Unlocking In-context Cross-lingual Performance

TL;DR

This paper introduces Cross-lingual QA, a prompting method that translates only questions and answers to improve cross-lingual performance in multilingual models, reducing translation costs and outperforming monolingual approaches.

Contribution

The paper proposes a novel cross-lingual prompting technique that translates only key parts of in-context examples, enhancing cross-lingual capabilities efficiently.

Findings

Cross-lingual QA outperforms monolingual prompting on diverse benchmarks.

Prompting with cross-lingual examples improves model performance as size increases.

The method reduces translation costs while maintaining contextual integrity.

Abstract

Multilingual large language models (MLLMs) have demonstrated significant cross-lingual capabilities through in-context learning. Existing approaches typically construct monolingual in-context examples, either in the source or target language. However, translating entire in-context examples into the target language might compromise contextual integrity and be costly in the case of long-context passages. To address this, we introduce Cross-lingual QA, a cross-lingual prompting method that translates only the question and answer parts, thus reducing translation costs. Experiments on four typologically diverse multilingual benchmarks show that Cross-lingual QA prompting effectively stimulates models to elicit their cross-lingual knowledge, outperforming prior monolingual prompting approaches. Furthermore, we show that prompting open-source MLLMs with cross-lingual in-context examples enhances performance as the model scale increases.