Cross-Lingual Retrieval Performance Under Varying High-to-Low Resource Language Ratios in Optimal Transport Distillation

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: What is the impact of varying the ratio of high-resource to low-resource languages in the training data on the cross-lingual retrieval performance of models fine-tuned with optimal transport distillation, as measured by nDCG@10 on MIRACL?

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