Impact of Hybrid Batch Sample Ratios on ALBEF Model Alignment Performance in CLIPScore Benchmark

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 varying the ratio of monolingual, cross-lingual, and multilingual samples in hybrid batch training affect the alignment performance of ALBEF models on the CLIPScore benchmark compared to models trained with uniform sample ratios?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.5/10.