Improving Zero-Shot Cross-Lingual Retrieval Robustness via Contrastive Learning on Code-Switched Data

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: Can the robustness of zero-shot cross-lingual retrieval models trained on code-switched data be improved by incorporating contrastive learning objectives, as evaluated by changes in MRR and nDCG scores on adversarial or noisy versions of the MIRACL benchmark?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.5/10.