Impact of Multilingual Data Proportion on Zero-Shot Cross-Lingual Retrieval Robustness in Low-Resource Settings

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: What is the impact of varying the proportion of multilingual data in hybrid batch training on the robustness of zero-shot cross-lingual retrieval across different language pairs in low-resource settings, as measured by recall@k on the Bucc-2018 dataset?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.5/10.