Impact of Artificially Code-Switched Training Data on Dense Retrieval Performance in Zero-Shot Cross-Lingual BEIR Benchmark

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: How does the use of artificially code-switched training data impact the performance of dense retrieval models on the BEIR benchmark compared to hybrid sparse-dense architectures in zero-shot cross-lingual settings?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.8/10.