Hybrid Batch Training for Robust Zero-Shot Cross-Lingual Retrieval

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: To what extent does the hybrid batch training strategy with multilingual contrastive learning objectives improve the robustness of zero-shot cross-lingual retrieval models on the BEIR benchmark when evaluated on adversarial or noisy queries, compared to baseline models measured by precision@5 and recall@20?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.1/10.