Performance Comparison of Multilingual Pre-trained Language Models in Cross-lingual Retrieval with 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: How does the performance of multilingual pre-trained language models compare on cross-lingual retrieval tasks when fine-tuned with optimal transport distillation versus traditional cross-entropy loss, evaluated on BEIR subsets for languages with fewer than 100 labeled training pairs?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 9.3/10.