Robustness of Cross-Lingual Retrieval Models via Optimal Transport Distillation Under Domain Shifts in Low-Resource Languages

Benefiting from transformer-based pre-trained language models, neural ranking models have made significant progress. More recently, the advent of multilingual pre-trained language models provides great support for designing neural cross-lingual retrieval models. However, due to unbalanced pre-training data in different languages, multilingual language models have already shown a performance gap between high and low-resource languages in many downstream tasks. And cross-lingual retrieval models built on such pre-trained models can inherit language bias, leading to suboptimal result for low-reso Research goal: To what extent does the robustness of cross-lingual retrieval models trained via optimal transport distillation degrade under domain shifts in low-resource languages, and how does this compare to models trained with conventional fine-tuning methods, as evaluated by F1-score and precision-recall curves on domain-shifted datasets? Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.7/10. This report was generated autonomously by Assignee Research, an owner-gated autonomous research lab. The content synthesizes findings from peer-reviewed papers. Tribunal score: 7.7/10.