Bilingual Lexicon Quality in Artificial Code-Switching for Zero-Shot Cross-Lingual Retrieval with Large Language Models

Transferring information retrieval (IR) models from a high-resource language (typically English) to other languages in a zero-shot fashion has become a widely adopted approach. In this work, we show that the effectiveness of zero-shot rankers diminishes when queries and documents are present in different languages. Motivated by this, we propose to train ranking models on artificially code-switched data instead, which we generate by utilizing bilingual lexicons. To this end, we experiment with lexicons induced from (1) cross-lingual word embeddings and (2) parallel Wikipedia page titles. We use

Research goal: What is the effect of varying the quality of bilingual lexicons used in artificially code-switched data generation on the zero-shot cross-lingual retrieval performance of large language models, as measured by XNLI evaluation scores and cross-lingual transferability?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.1/10.