Scaling of Zero-Shot Cross-Lingual Retrieval Performance with In-Language Training Data on MTOP

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 zero-shot cross-lingual retrieval performance of the hybrid batch training approach scale when incorporating progressively larger amounts of in-language training data on the MTOP benchmark, measured by nDCG@10 across varying language pairs?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.2/10.