Hybrid Batch Training on XQuAD: Scaling Multilingual Pairs and the Zero-Shot versus Monolingual Accuracy Trade-off

Information retrieval across different languages is an increasingly important challenge in natural language processing. Recent approaches based on multilingual pre-trained language models have achieved remarkable success, yet they often optimize for either monolingual, cross-lingual, or multilingual retrieval performance at the expense of others. This paper proposes a novel hybrid batch training strategy to simultaneously improve zero-shot retrieval performance across monolingual, cross-lingual, and multilingual settings while mitigating language bias. The approach fine-tunes multilingual lang

Research goal: How does the hybrid batch training approach perform on the XQuAD benchmark when scaled with additional multilingual language pairs beyond the 11 languages in TyDi QA, and what is the trade-off between zero-shot cross-lingual retrieval accuracy and monolingual accuracy?

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