Generalization of Hybrid Batch Training to Multilingual Retrieval Benchmarks Beyond MIRACL

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: Does the proposed hybrid batch training strategy in the paper generalize to other multilingual retrieval benchmarks beyond MIRACL, such as BEIR or XTRT, and how does it compare to standard multilingual fine-tuning in terms of cross-lingual zero-shot accuracy?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.7/10.